DANETV
Baltelie
ine lUintiiJM ^ Innm-d
LUMIStox 300 Bench Top Luminometer
ECLOX Handheld Luminometer
Joint verification report
Luminescent bacteria test for use in wastewater
Handheld ECLOX
LUMIStox 300
May 2011
Final
-------
LUMIStox 300 Bench Top Luminometer
ECLOX Handheld Luminometer
Joint verification report
Agern Alle 5
DK-2970 H0rsholm
Denmark
Tel: +4545169200
Fax: +4545169292
mta@dhigroup.com
www.dhigroup.com
Vendor
HACH-LANGE GmbH
Vendors representative
Dr. Elmar Grabert
Projecl
DANETV
Authors
2
1
0
Revision
Mette Tjener Andersson
Claus J0rgensen
Final report
Verification report for second review
Verification report for review
Description
Key words
Acute toxicity; EC50; ECLOX; ISO 11348-3;
Luminescent bacteria; LUMIStox; Wastewater
Project No
11800378
Date
May
Approved
2011
by
Berislav Tomicic
CLJ
CLJ
MTA
By
MWN BE
MWN HC
MWN HC
Checked Appr
ET 20113005
3E 20112402
3E 20100604
oved Date
Classification
E3 Open
n Internal
n Proprietary
Distribution
HACH-LANGE GmbH:
Battelle:
US-EPA:
ETV Canada:
DHI:
Dr. Elmar Grabert
Mary Schrock
John McKernan
Mona EI-Hallak
MTA-MWN-CLJ-BOP
No of copies
1+file
1+file
1+file
file
1+file
-------
TABLE OF CONTENTS
1 INTRODUCTION 1
1.1 Name of product 1
1.2 Name and contact of vendor 1
1.3 Name of center/verification responsible 1
1.4 Verification test organization 2
1.5 Technical experts 3
1.6 Verification process 3
2 DESCRIPTION OF THE TECHNOLOGY 4
3 DESCRIPTION OF THE PRODUCTS 6
3.1 LUMIStoxSOO 6
3.2 ECLOX 6
4 APPLICATION AND PERFORMANCE PARAMETER DEFINITIONS 7
4.1 Application definition 7
4.2 Performance parameters for verification 7
4.3 Additional parameters 9
5 EXISTING DATA 10
5.1 Summary of existing data 10
5.2 Quality of existing data 10
5.3 Accepted existing data 10
6 TEST PLAN REQUIREMENTS 11
6.1 Test design 11
6.2 Comparable tests and chemical analysis 12
6.3 Data management 13
6.4 Quality assurance 13
6.5 Test report 14
7 EVALUATION 15
7.1 Calculation of performance parameters 15
7.2 Performance parameter summary 17
7.2.1 Criterion of detection 17
7.2.2 Range of application 17
7.2.3 Precision 18
7.2.4 Agreement with accepted values 19
7.2.5 Robustness 20
7.3 Evaluation of test data quality 25
7.3.1 Reference chemical analysis performance data 25
7.3.2 Comparable test performance data 26
7.3.3 Test system control data 26
7.3.4 Audits 30
7.3.5 Deviations 30
7.4 Additional parameters summary 30
7.4.1 User manual 30
-------
7.4.2 Product costs 32
7.4.3 Occupational health and environment 33
7.5 Operational parameters 33
7.6 Recommendation for verification statement 34
8 VERIFICATION SCHEDULE 38
9 QUALITY ASSURANCE 39
APPENDIX 1 40
Terms and definitions used in the verification protocol 40
APPENDIX 2 44
References 44
APPENDIX 3 48
Application and performance parameter definitions 48
APPENDIX 4 58
Deviation report for verification and testing 58
-------
1 INTRODUCTION
Environmental technology verification (ETV) is an independent (third party)
assessment of the performance of a technology or a product for a specified application,
under defined conditions and quality assurance.
This verification is a joint verification between Danish Centre for Verification of
Climate and Environmental Technologies (DANETV), the U.S. Environmental
Technology Verification (U.S. EPA ETV) Advanced Monitoring Systems (AMS)
Center and the Canadian ETV Program (ETV Canada). The objective of the
verification was to evaluate the performance of a wastewater rapid toxicity technology
that can be used to monitor industrial or domestic wastewater.
1.1 Name of products
The verification report covers two products from the same vendor; both are acute
toxicity tests with luminescent bacteria. The target products were LUMIStox 300
bench top luminometer and ECLOX handheld luminometer. Both can operate in
connection with a LUMIStherm thermostat and the PC software LUMISsoft4, version
2.0.2.56 (December 2009).
1.2 Name and contact of vendor
HACH-LANGE GmbH, Willstatterstrasse 11, 40549 Dusseldorf, Germany, phone +49
211 52880.
Contact: Dr. Elmar Grabert, email: elmar.grabert@hach-lange.de, phone +49 211 5288
241.
Web site: www.hach-lange.de
1.3 Name of center/verification responsible
Danish Centre for Verification of Climate and Environmental Technologies,
(DANETV), DHI DANETV Water Centre, DHI, Agern Alle 5, DK-2970 H0rsholm,
Denmark.
Verification responsible: Mette Tjener Andersson, email mta@dhigroup.com,
phone+45 16 91 48.
U.S. EPA ETV Advanced Monitoring Systems Center (Battelle), Battelle Memorial
Institute, 505 King Avenue, Columbus, Ohio 43201-2693, U.S.A.
Verification Test Coordinator: Mary E. Schrock, email schrock@battelle.org,
phone+1 6144244976.
ETV Canada, 2070 Hadwen Road Suite 201 A, Mississauga, Ontario L5K 2C9,
Canada.
Verification responsible: MonaEl Hallak, email melhallak@etvcanada.ca,
phone +1 905 822 4133 extension 239.
-------
1.4 Verification test organization
The verification was conducted as a joint verification between the DANETV, the U.S.
EPA ETV program and ETV Canada. The verification was planned and conducted to
satisfy the requirements of the ETV scheme currently being established by the
European Union (EU ETV) as well as the U.S. and Canadian ETV programs.
Verification and tests were performed by DHI as DANETV Water Technology ETV
Center (DHI DANETV Water Centre) under contract with the Danish Agency for
Science, Technology and Innovation. Battelle participated as the manager of the ETV
AMS Center through a cooperative agreement with the U.S. Environmental Protection
Agency (EPA). ETV Canada participated as manager of the Canadian ETV Program.
The day-to-day operations of the verification and tests were coordinated and
supervised by DHI personnel, with participation of the vendor, HACH-LANGE. The
testing was conducted in DHI laboratories, H0rsholm, Denmark. DHI personnel
operated the luminometers during the verification. HACH-LANGE provided
luminometers, thermostats, bacteria, software, user manuals and operation instructions.
HACH-LANGE furthermore participated in development of protocol and plans by
providing input to DHI. Battelle and ETV Canada ensured that the verification and
tests were planned, conducted and reported to satisfy the requirements of the U.S. and
Canadian ETV programs, including input and concurrence from their stakeholder
groups, as described in the process document III. Battelle and ETV Canada also
participated in the development of the verification protocol, test plan, verification
report, and verification statement and they performed quality assurance (QA) of the
verification and tests. The verification protocol, test plan, test report, verification
report, and verification statements were reviewed and approved by ETV Canada, while
U.S. EPA ETV AMS Center and Environment Canada reviewed and approved all
listed documents except the test report.
Three technical experts provided independent expert reviews of the planning
documents. Four experts provided reviews of the verification report. The test report is
solely a DANETV report; DANETV requires review by one external expert. The test
report was therefore reviewed by only one of the external experts.
The chart in Fig 1.1 identifies the relationships of the organizations associated with this
verification and test.
-------
U.S.EPAETV Environment DANETV
Canada ETV
Program
AMS Center
(Battelle) ETV Canada Verification
DHIWTC
Test
ETV AMS Center ETV Canada
Stakeholders Stakeholders
Technical experts HACH-LANGE
Fig 1.1 Organization of the verification and tests.
1.5 Technical experts
The technical experts are:
Dr. Joel Allen, email: allen.joel@epa.gov, phone +1 513 487 2806. U.S.EPA, Office of
Research and Development/National Risk Management Research Laboratory /Water
Supply and Water Resources Division/Water Quality Management Branch.
Associate Professor Kresten Ole Kusk, email: kok@env.dtu.dk, phone +45 4525 1569.
Technical University of Denmark, Department of Environmental Engineering.
Dr. Ali Safarzadeh-Amiri, email: Amiri.s.ali@gmail.com, phone +1- 905-827-7859.
Amiri Clean Water Technologies, Oakville, Ontario, Canada, L6M 4W5.
Dr. Max Lee, email: mmlee@dow.com, phone +1 979 238 7726. Environmental Tech
Center, Dow Chemical Company.
1.6 Verification process
The principles of operation with the role of the verification and test documents and the
different sub-bodies responsible are given in Fig 1.2.
-------
Fig 1.2
Principles of operation of the DANETV verification scheme for joint verification.
The QA group covers the expert group, Battelle, U.S. EPA ETV and ETV Canada.
Audits were performed internally by DHI and Battelle for U.S. EPA ETV.
References for the verification process were the Quality Management Plan from
Battelle 121, the General Verification Protocol from ETV Canada /3/ and the Quality
Manual for the ETV operations at DHI following the DANETV Quality Manual
Template /4/.
The final verification protocol, the test plan, and the above mentioned process
document were the planning documents for this verification test.
Two separate joint verification statements, one for each product, were issued after
completion of the verification. The results of verification and testing were described in
one verification report and one test report covering both the LUMIStox 300 Bench Top
Luminometer and the ECLOX Handheld Luminometer.
DESCRIPTION OF THE TECHNOLOGY
Luminometers such as LUMIStox and ECLOX are in vitro testing systems that use
bioluminescent bacteria to detect toxic compounds in water. Luminometers can directly
determine toxicity of water soluble chemicals, and from a number of compatible water
matrices such as river, lake and wastewater, and leachates from soil, waste or rubble.
Bioluminescence tests are metabolic inhibition tests that provide acute toxicity
analyses. For the LUMIStox and ECLOX technologies, a strain of naturally occurring
luminescent bacteria, Vibrio fischeri, is used. Vibrio fischeri is a non-pathogenic,
marine, luminescent bacterium which is sensitive to a wide range of toxicants and is
commonly used in rapid toxicity tests. When properly grown, luminescent bacteria
produce light as a by-product of their cellular respiration. Any inhibition of cellular
activity results in a decreased rate of respiration and a corresponding decrease in the
rate of luminescence. For this verification, the light emission/luminescence was
measured with a LUMIStox or ECLOX luminometer.
-------
Inhibition of the light emission in the presence of a sample is determined relative (as
percent inhibition) to a non-toxic control. The luminescence is measured after a contact
time of five (optional), 15 and 30 minutes at 15 °C, taking into account a correction
factor, which is a measure of the control sample's intensity change during the exposure
time.
-------
3 DESCRIPTION OF THE PRODUCTS
3.1 LUMIStox 300
The LUMIStox 300 (referred to as "LUMIStox" throughout this report) is a bench top
luminometer that has been developed as a measuring unit for the luminescent bacteria
test. In combination with the LUMIStherm incubation block, it conforms to the
technical requirements of ISO 11348. This ISO standard describes determination of the
inhibitory effect of water samples on the light emission of Vibrio fischeri. The ISO
standard contains three parts, using freshly prepared bacteria, liquid-dried bacteria and
freeze-dried bacteria, respectively. For the LUMIStox (and ECLOX) freeze-dried
bacteria are used. Therefore ISO 11348-3 7237 applies.
The LUMIStox 300 has a built-in photometer function and an automatic measuring and
evaluation routine, which enables it to recognize color effects in the luminescent
bacteria test, and to take these into account in the test results.
The photometer function also allows the color effect to be estimated in advance, and
can be used to determine the extinction (as OD - optical density) of bacteria
suspensions for the purpose of assessing light extinction.
The LUMIStox 300 can be connected to a personal computer running the LUMISsoft4
that enables the operator performing and recording luminescent bacteria tests to
conduct all of the ISO 11348-3 requirements. The results from the measurements are
percent inhibition, but with use of the software LUMISsoft4 either Lowest Ineffective
Dilution (LID) or Effective Concentration (EC) values, representing concentrations
causing less than 20%l inhibition can be determined. EC values can be extrapolated to
concentration values causing 50% inhibition (ECso) using a model not validated in this
verification. ECso values are the commonly used results from toxicity tests
internationally, while the LID is used as a standard practice in Germany.
3.2 ECLOX
The ECLOX is a portable instrument designed to provide data appropriate for risk
assessments in the event of environmental releases, emergency situations, preventive
security measures, and regulatory monitoring.
The ECLOX is designed in particular to be used for the luminescent bacteria toxicity
test and to be used with a chemiluminescence toxicity test. Both tests will give quick
results in the field or in the laboratory. The ECLOX used in the field provides values of
percent inhibition.
Additionally, the ECLOX can be used in the laboratory in the same way as the
LUMIStox. When the ECLOX is used with the thermal block LUMIStherm and
connected to a PC with the software LUMISsoft4, the principles of the luminescent
bacteria test according to ISO 11348 can be followed (however, tests performed on the
ECLOX are not ISO 11348-3 compatible). For the LUMIStox, the percent inhibition
results can be used to calculate LID and ECso values.
1 LID of 20% inhibition is stated in ISO 11348-3, Annex B, Section B.5.
-------
4 APPLICATION AND PERFORMANCE PARAMETER
DEFINITIONS
The application has been defined in detail in Appendix 3 for matrices for use, targets of
monitoring, and effects. The application and performance parameters are summarized
in this section.
4.1 Application definition
An overview of matrix, effect, targets and technologies is given in Table 4.1.
Table 4.1 Description of matrix, effect, targets and technologies.
Matrix
LUMIStoxand
ECLOX are applied
for wastewater; river
and lake water;
leachate from soil,
waste, rubble, etc.;
or directly in fluent
chemicals.
Verification testing
was conducted on
domestic and
industrial wastewater
effluents.
Effect
Measurement of
toxicity as indicated
by inhibition of
luminescent bacteria
by a variety of
compounds including
metal ions, organic
pesticides, inorganic
and organic
pollutants and
surfactants.
Additional
parameters:
User manual quality,
product cost,
environmental health
and safety.
Targets
The target for the
application is
measurement of
toxicity, specifying
criterion of detection
(CD), range of
application, precision
(repeatability and
reproducibility),
agreement with
accepted values and
robustness.
Technologies
ECLOX and
LUMIStox analyses
for inhibition of light
emitting
luminescent
bacterium Vibrio
fischeri.
4.2 Performance parameters for verification
The performance parameters relevant for the application, as derived in Appendix 3, are
presented in Table 4.2. The ranges presented for these parameters were used for
planning the verification and testing only and will not be compared to actual
performance.
Table 4.2 Relevant ranges of performance parameters in effluent industrial and domestic wastewater.
LUMIStox
ECLOX
Criterion
of
detection
% inhibition
<10
<10
Range
of
application
Dilution L/L
>1/2-< 1/32
>1/2-< 1/32
Precision (RSD)
%
Repeatability
<20
<20
Reproducibility
<30
<30
Agreement
with
accepted
values
%
100±50
100±50
Robust-
ness
%
100±50
100±50
-------
For toxicity testing, it is not possible to determine the limit of detection (LoD). Instead,
a criterion of detection (CD) was chosen, above which inhibition is seen as significant,
based on the standard deviation of blanks (2% NaCl solution and bacteria suspension,
no toxic compound added).
The range of application for a chemical analysis is usually the range of analyte
concentration from the limit of detection to the highest concentration with linear
response. This concept is not meaningful for a toxicity test of a water sample, because
the test does not measure a concentration but an inhibitory effect as a function of the
dilution of the sample. The range of application for determining ECso therefore has to
be considered in terms of dilution. According to the HACH-LANGE manual,
estimation of ECso for a water sample requires a minimum of three measurements
where the inhibition is between 10% and 90%. In addition one of the three
measurements must be above 50%. If the standard dilution row is considered as
described in the LUMIStox 300 operation manual and in Annex B of the ISO 11348-
3:2007 with nine dilutions (2, 3, 4, 6, 8, 12, 16, 24, 32 times dilution in the test
suspension), then ECso should be in the range of dilutions greater than two and less
than 32 times. This assumes three measurements with inhibition between 10 and 90%.
Based on test results, ranges of concentrations of the compounds tested should give
inhibition within the range of application. The range of application will be given in mg
compound/L and is valid for undiluted samples. If samples are more toxic than the
maximum value in the range of application, additional dilution shall take place prior to
testing. If samples are less toxic, a minimum value in the range of application (ECso
values) cannot be determined.
Precision can be evaluated under repeatability and reproducibility conditions.
Repeatability is defined as the relative standard deviation of measurements done with
the same measurement procedure, operators, measuring system, operating conditions,
and location with replicate measurements on the same or similar objects over a short
period of time. Reproducibility is defined as the relative standard deviation (RSD) of
measurements under different conditions such as locations, operators, and measuring
systems with replicate measurements on the same or similar objects. In laboratory
terminology, repeatability is the within-series precision and the reproducibility the
between-series precision. For reproducibility of luminescent toxicity testing, the
difference in bacteria batches is considered to be the greatest source of deviation and is
one of the variables which were evaluated in this verification. The other variables were
different days and different technicians. Precision has been determined as the RSD of
the EC20 and ECso results generated during testing.
"Trueness" is the closeness of agreement between the (mean) concentrations found in
measurements, and the true or accepted concentration. According to ISO 11348-3 the
true or accepted ECso value of a substance is obtained, as long as the criteria in the ISO
method are met. For this verification it was chosen to determine trueness as
"agreement with accepted values." This agreement is the inhibition results (ECso
values) obtained in the tests compared to robust literature values for ECso values, with
clear reference to tests performed according to the ISO 11348-3 method for the same
compound. The agreement with accepted values was only determined for test
substances where robust literature values were available.
The verified parameters for "robustness" included pH change, temperature change,
presence of color or turbid material in the sample, difference in initial concentration
8
-------
(i.e., lowest dilution of the sample), matrix variation, and type of cuvette. Robustness
was the trueness for each of the verified parameters.
Samples were tested with different concentration of color and turbid material, since the
ISO standard specified this would cause interference. Available color correction
methods were used for both the LUMIStox and ECLOX during the verification.
The ISO 1 1348-3 recommends testing be performed at a pH range of 7.0±0.20, but
stated that pH values of 6.0-8.5 are acceptable. Tests were performed comparing three
pH values (6.0, 7.0, and 8.5).
The ISO 1 1348-3 specifies that a thermostat should be used to cool the test vials to
15±1 °C. A monitored thermostat was used during the verification testing. Tests were
performed comparing temperatures of 14.0 °C, 15.4 °C and 16.1 °C.
When testing wastewater samples, it is not always possible to cover the ideal range
from 10 to 90% inhibition. Tests were therefore performed with maximum
concentrations of approximately 30% and 60% inhibition (ECso and ECeo), to see how
that affected the determination of £€20 and ECso. Initial concentrations causing
approximately 30% and 60% inhibition were used to determine EC 20. Initial
concentrations causing 60% inhibition were used to determine
Testing of industrial and domestic effluent wastewater samples was included. This
included testing of these wastewaters as they were received. To show they were non-
inhibitory, these water samples were tested with and without spiking using inhibitory
chemicals. These tests were performed to evaluate the effect of the wastewater matrix
on the luminescent test.
Typically glass cuvettes are used in the LUMIStox, and plastic cuvettes are used in the
ECLOX. HACH-LANGE has stated that plastic cuvettes can also be used in
LUMIStox. To be consistent, all tests were performed with plastic cuvettes except for
test L, where the LUMIStox was tested for robustness using both types of cuvettes
(glass and plastic).
4.3 A dditional parameters
Besides the performance parameters obtained by testing, a compilation of parameters
describing the ease of understanding the user manual, product costs, and occupational
health and safety issues of the product were included in the verification.
-------
5 EXISTING DATA
5.1 Summary of existing data
The vendor recently performed tests with the LUMIStox and ECLOX instruments for
determination of precision expressed by the relative standard deviation (RSD). Table
5.1 provides results from HACH-LANGE at a contact time of 15 minutes.
Table 5.1 Results from testing performed by HACH-LANGE of LUMIStox and ECLOX.
Compound
CrB+
Zn^
Pb^
SDS1
CTAB2
Formaldehyde
Hydroquinone
p-Cresol
CN"
Range 10-90%
inhibition
mg/L
1.7-27
1.5-9.0
0.21-2.5
0.14-2.3
0.33-6.0
4.4-35
0.03-0.20
0.38-6.0
0.51-8.1
LUMIStox
No. of bacteria
batches/no, of
replicates
3/5
2/4
2/4
3/6
2/4
2/4
2/7
2/4
2/6
EC50
mg/L
6.6
4.3
0.49
0.66
0.84
15
0.09
1.5
2.7
RSD
%
38
25
8.0
16
5.8
9.5
46
33
74
ECLOX
No. of bacteria
batches/no, of
replicates
1/3
1/3
1/3
1/3
1/3
1/3
EC50
mg/l
8.6
4.2
0.48
0.55
1.1
14
RSD
%
26
15
8.7
2.8
16
5.1
Not tested
1/3
1.6
6.6
Not tested
1: Sodium Lauryl Sulphate.
2: Cetyl Trimethyl Ammonium Bromide.
The range 10% to 90% inhibition was the measurement interval used for calculating
the ECso values. Ten percent inhibition equals ECio, while 90% inhibition equals ECgo.
This range for compounds was used as guidance for the test range included in the
verification.
It should be mentioned that the RSD was calculated by the vendor as a general RSD
including all results, and with no reference to number of samples tested in each
bacteria batch. Note that the test of LUMIStox was performed on two to three different
bacteria batches, while the test of ECLOX was performed on one bacterial batch only.
This resulted in higher RSDs for LUMIStox as compared to ECLOX.
The vendor made a note on results regarding cyanide being difficult to work with in the
laboratory at a pH =7.
At pH = 7, almost all cyanide is in the volatile and toxic hydrogen cyanide (HCN)
form, and evaporation of HCN can occur.
5.2 Quality of existing data
The tests were performed by the vendor, and not by an independent body. Furthermore,
the analyses were not conducted by a laboratory with ISO 17025 accreditation.
5.3 Accepted existing data
No existing data were accepted for use as part of the verification test. However, these
data did provide useful background for planning the test.
10
-------
6 TEST PLAN REQUIREMENTS
Based upon the application and performance parameters identified in Section 4, the
requirements for test design were established in the test plan. The detailed test plan was
prepared separately, based upon the test requirements summarized below.
6.1 Test design
The outline of the required tests is shown in Table 6.1. More details of the test design
can be found in the test report /30/. The principle behind the design was that three test
set-ups were used:
• LUMIStox 300 bench top luminometer with LUMIStherm thermostat and
LUMISsoft4 PC software. According to ISO 11348-3.
• ECLOX handheld luminometer with LUMIStherm thermostat and LUMISsoft4 PC
software. Conditions similar to ISO 11348-3.
• ECLOX handheld luminometer with use of firmware.
Three matrices were used in the testing: spiked 2% sodium chloride (NaCl) MilliQ
water, domestic effluent wastewater, and industrial effluent wastewater. Salinity of the
wastewaters was increased to 2% by addition of solid NaCl.
Tests were performed with specific compounds in 2% NaCl MilliQ water to determine
their EC20- and ECso values. The tests showed the range of responses towards these
specific toxic compounds (zinc (Zn +), chromium (Cr2O72-), triclosan, cyanide (CN-),
sodium lauryl sulphate (SDS) and cetyl trimethyl ammonium bromide (CTAB)).
Secondly, tests were performed on effluent wastewater with and without spiking with a
toxic compound. This showed the robustness of the luminescent tests towards the
wastewater matrix. The last test evaluated the effect on results between use of glass
cuvettes and plastic cuvettes in the LUMIStox Benchtop.
11
-------
Table 6.1 Test design and associated performance parameters.
Test
no.
A
B
C
D
E
F
G
H
I
J+K
L
Performance parameters
Range, Repeatability, Agreement with accepted values
Criterion of detection
Robustness, effect of start cone, on repeatability
Reproducibility
Robustness, sample temperature at field use
Robustness, sample temperature at laboratory use
Robustness, pH
Robustness, color
Robustness, turbidity
Robustness, matrix
Robustness, cuvettes
Equipment
LUMIStox
X
X
X
X
X
X
X
X
X
X
ECLOX incl.
thermostat and
software
X
X
X
X
X
X
X
X
X
ECLOX incl.
firmware
X
Matrix
g
i
o
re
z
sS
cs
X
X
X
X
X
X
X
X
X
X
Waste water
X
ISO 11348-3 requires that each batch of bacteria is tested by determining the inhibition
by three reference substances. These tests were performed solely on the LUMIStox,
since the operation of the ECLOX is not in compliance with the ISO 11348-3.
6.2 Comparable tests and chemical analysis
Reference tests conducted by an independent laboratory using Vibrio fischeri following
ISO 11348-3 were originally planned. However, the limited utility of the planned
reference tests was noted in the Joint Verification Protocol 7337 and in the Joint Test
Plan 7347.
The reference tests were intended to be done under ISO 17025 accreditation, using the
ISO 11348-3 luminescent bacteria test method with Microtox equipment. ALcontrol
was selected as the independent laboratory to conduct the ISO 11348-3 accredited
testing. The results obtained by ALcontrol for one of the reference compounds were
lower than anticipated, that is the control compound appeared more toxic than
anticipated (see section 7.3.3). The systems audit in section 7.4 identified that
ALcontrol was accredited to conduct ISO 11348-3, but used a modified method. After
scrutinizing the first set of results from the laboratory, and after subsequent discussions
with them, the data impact of the modified ISO 11348-3 method were realized. Hence,
it was determined that the comparison of ALcontrol data to the HACH-LANGE results
would be of lower value, since the two methods were not directly comparable. It was
investigated if the tests could be performed elsewhere fulfilling the ISO 11348-3 and
the accreditation requirement as well as operating different equipment than the
LUMIStox (or ECLOX). A laboratory meeting these criteria could not be found in
Germany, where the HACH-LANGE equipment is widely used, or in Norway.
Laboratories were found that conducted a modified version of the ISO 11348-3
method, but none were found that would conduct the ISO method as written.
Therefore, it was decided to exclude further reference tests rather than include
measurements from a laboratory that was performing a modified method, since the
12
-------
results would likely be different. This change to the test plan was documented in
Deviation 9B (see Appendix 4). Precedence for not using a reference test for water
toxicity verification testing followed the U.S. ETV Test/QA Plan for Verification of
Rapid Toxicity Technologies 7357. DHI believes the impact of not including these
analyses was minimal, since these data were not intended to be used as true reference
measurements but rather to present results that would have been obtained by a
comparable technology. Additionally, data from peer-reviewed scientific literature
based on ISO 11348-3 had been planned for comparison to the testing data, and are
available (see section 7.2.4).
Reference chemical analyses of stock solutions were done under ISO 17025
accreditation 7177 with appropriate methods by an independent laboratory.
6.3 Data management
Data storage, transfer and control were done in accordance with the requirements of
ISO 90017187 enabling full control and retrieval of documents and records. The filing
and archiving requirements of the DHI Quality Manual were followed (10 years
archival).
6.4 Quality assurance
The quality assurance (QA) of the tests included audits of the test system at DHI
DANETV Water Centre and the external laboratory performing reference tests, as well
as performance evaluations of the laboratory providing stock solution confirmations.
Data quality audits were performed on data generated during testing to ensure data
quality and integrity.
This verification report was subjected to review by the QA group indicated in Fig 1.2.
Since this verification was a joint verification with the U.S. EPA ETV and ETV
Canada, an on-site technical systems audit (TSA) by the Battelle AMS Center was
included as part of the quality assurance. An audit debrief occurred at the conclusion of
the TSA, and issues identified during the audit were brought to DHI's attention. This
included issues which were ultimately identified as one finding, four observations, and
one recommendation. The finding raised during the TSA debrief was that the external
laboratory was performing a modified ISO 11348-3. At the time of the TSA, the
impact of the modifications on data quality was not known. However, DHI further
investigated the external laboratory upon receipt of the first batch of data, and
determined that the modifications had impact on usability of the data. As a result, the
use of modified ISO 11348-3 was discontinued and documented as a finding in the
final audit report.
The Battelle Quality Manager and the ETV Canada Quality Manager also performed
an audit of data quality. This was a review of data acquisition and handling procedures
and an audit of at least 10% of the data acquired in the test and verification. The
Quality Managers traced the data from initial acquisition, through reduction and
statistical comparisons, to final reporting. All calculations performed on the data
undergoing the audit were checked.
13
-------
6.5 Test report
The test report /30/ followed the principles of the template of the DHIDANETV
verification center quality manual template 141 with data and records from the tests
presented.
The test report was not reviewed by the U.S. ETV program or the Battelle AMS
Center, since the purpose of the test report was a specific requirement for DANETV.
One test report was prepared for both verified technologies (LUMIStox and ECLOX).
14
-------
7 EVALUATION
The evaluation included calculation of the performance parameters from Section 4.2,
evaluation of the data quality based upon the test quality assurance from Section 6.4,
and compilation of the additional parameters from Section 4.3.
The calculations involved in the £€20 and ECso determination by the LUMISsoft4
software were not independently verified as part of this test. However, results
generated by the software were spot-checked by comparison to calculations derived
independently, e.g. when performing manual color correction of results from the
ECLOX and calculating EC values.
7.1 Calculation of performance parameters
By testing a dilution series with inhibitions in the range from 10%-90%, EC2oand ECso
values can be calculated according to principles in ISO 11348-3. This is performed by
the software LUMISsoft4 connected to the HACH-LANGE instruments. To estimate
ECso values, a minimum of three measurements have to be in the range from 10%-90%
inhibition. Furthermore, one concentration has to give response above 50% inhibition
of a valid ECso value.
For use of the ECLOX without connection to a computer, the results were recorded as
percent inhibition and, as such EC values could not be determined directly.
Calculations of parameters and EC values (and in the case of ECLOX using firmware,
percent inhibition) were performed according to accepted statistical principles (Table
7.1 and 191). Table 8.1 includes updates to the calculations originally listed in the
verification protocol 7337 that were added to improve the quality of the evaluation, and
are described in Deviation 8 (see Appendix 4).
15
-------
Table 7.1 Calculations used for the test results
Parameter
Calculation
Explanations
Criterion of detection
CD is criterion of detection;
to.gsff) is the Student's t factor for f where f= n-1
degrees of freedom;
n is number of measurements;
Skis a pooled estimate for standard deviation of
luminescent in control glasses
Range of application
Minimum: just above 2*ECso
Maximum: just less than 32*ECso
ECso: Concentration causing 50% inhibition
Precision (repeatability),
as relative standard
deviation, RSD
; max ; min
n
d =
m
DI is the range at level i;
Ximin and ximax are the lowest and highest
measurements at level i;
Xi is the average of n measurements;
m is the number of levels;
di is the relative range at level i;
^ d is the mean relative range for all m levels
Divisor is for i=3 equal to 1.693 and for i=4 equal to
2.059
Divisor
Precision
(reproducibility), as
relative standard
deviation, RSD
n
^Between groups I M^Between grou
Xi is the average of n measurements in group;
m is the number of levels;
s is standard deviation
F, is average of average in groups;
s is standard deviation
MSeetween groups is variance between groups obtained
by single factor ANOVA in Excel
RSD^SBetween groups ^WQ%
Agreement with
accepted values, A.
Based on robust
literature values
(obtained by use of ISO
11348-3)
n
n
xt is the mean of measurements at level i, Xj;
y t is the literature value at level i, yi;
Ai is the agreement at level i;
A is the mean agreement for all levels
= — x 100 %
Robustness, R
R= £ x 100
xre
xro is the average of measurement under
conditions of robustness test;
xre is the average of measurements under
reference conditions
Test of significant
deviation from
reference.
Used for robustness
results
sx is standard deviation on dataset x
sy is standard deviation on dataset y
fx is degree of freedom for dataset x
fy is degree of freedom for dataset y
x is the average of measurements of dataset x
y~is the average of measurements of dataset y
sxy is average deviation
nx is number of measurement in dataset x
ny is number of measurement in dataset y
to.gys is student t-factor for two-sided test
16
-------
Calculations of performance parameters were conducted in Excel 2007.
7.2 Performance parameter summary
Results in the test report /30/ are given for a test time of both 15 and 30 minutes. When
EC values are calculated both EC20 and ECso values are listed. In the verification report
only ECso values are listed, since they are most widely used. In the verification
statement, only results for ECso with a test time of 30 minutes are listed, since this test
time is used in reporting more frequently than the 15-minute test times.
7.2.1
Criterion of detection
The criterion of detection, the level above which inhibition is significant (95%), was
calculated based on series of nine 2% MilliQ water samples including bacteria, but no
toxic compounds. The criteria of detection for LUMIStox and ECLOX after 15 and 30-
minute exposures, respectively, are given in Table 7.2.
Table 7.2 Criterion on detection (% inhibition). Number of replicates (n) is 3.
Test time
(min)
15
30
LUMIStox
(% inhibition)
6.7
5.8
ECLOX
(% inhibition)
7.5
5.5
7.2.2 Range of application
Range of application in this context means the concentration range where (pure water)
samples can be tested without dilution or pre-concentration.
Table 7.3 LUMIStox range of application in 2% NaCI MilliQ water for target compounds (mg/L).
Number of replicates (n) is 3 and 4 for cyanide.
LUMIStox
Compound
Znz+
Cr20/-
Triclosan J
Cyanide
SDSJ
CTABJ
15 min
Average
EC50
(mg/L)
8.5
n.c. n
0.40
24 z
1.4
1.3
Range of application
(mg/L)
Minimum
>17
-
>0.79
>48
>2.8
>2.7
Maximum
<270
-
<13
<770
<44
<43
30 min
Average
EC50
(mg/L)
4.1
17
0.53
24
1.0
0.97
Range of application
(mg/L)
Minimum
>8.3
>35
>1.1
>48
>2.0
>1.9
Maximum
<130
<560
<17
<780
<32
<31
n.c.: Not calculated.
1 EC50 for Cr2O72" was not possible to calculate after 15 minutes. The requirement of one measurement above 50%
inhibition was not fulfilled.
2 EC50 for cyanide was only possible to calculate after 15 minutes for two out of four replicates. The requirement of one
measurement above 50% inhibition was not fulfilled.
3 The recovery of these compounds in mixed solutions was not near 100%. The listed EC values are based on the
added amount of compound. See details on recovery later in section 7.3.3.
The range of application was based on ECso values determined for six target
compounds. Note that originally the verification protocol /33/ called for using nine
target compounds; however, three of these compounds: CuSC>4 (heavy metal),
Flutriafol (organic pesticide), and 4-NPE (surfactant) were not sufficiently toxic at
concentrations without precipitation to be used for testing and were therefore excluded.
17
-------
Removal of these three target compounds is described in Deviations 1, 2 and 4,
respectively (see Appendix 4). The six remaining target compounds represent the
compound categories of heavy metals, organic compounds, industrial pollutants, and
surfactants.
Table 7.4 ECLOX range of application in 2% NaCI MilliQ water for target compounds (mg/L). Number
of replicates (n) is 3 and 4 for cyanide.
ECLOX
Compound
Zn2+
Cr2072-
Triclosan 4
Cyanide
SDS4
CTAB4
15 min
Average
EC50
(mg/L)
8.4
n.c.1
0.39
23 3
1.4
1.4
Range of application
(mg/L)
Minimum
>17
-
>0.77
>45
>2.8
>2.9
Maximum
<270
-
<12
<730
<45
<46
30 min
Average
EC50
(mg/L)
4.1
18
0.53
18
0.99
0.96
Range of application
(mg/L)
Minimum
>8.2
>37
>1.1
>35
>2.0
>1.9
Maximum
<130
<590
<17
<570
<32
<31
n.c.: Not calculated.
1 EC50 for Cr2O72" was not possible to calculate after 15 minutes. The requirement of one measurement above 50%
inhibition was not fulfilled.
2 EC50 for cyanide was only possible to calculate after 15 minutes for three out of four replicates. The requirement of
one measurement above 50% inhibition was not fulfilled.
3 EC50 for cyanide was only possible to calculate after 30 minutes for three out of four replicates. The requirement of
one measurement above 50% inhibition was not fulfilled.
4 The recovery of these compounds in mixed solutions was only 2-7%. The listed EC values are based on the addition
of compound. See details on recovery in section 7.3.3.
To be able to determine the ECso value, an initial concentration greater than twice the
ECso is needed, since the standard procedure is to dilute the sample to half the initial
concentration before testing. Without extraordinary dilution of the sample, the ECso
value has to be detected within the regular dilution series containing nine dilutions
(limitation by the thermoblock). The maximum concentration in the sample can
therefore be less than 32 times the ECso. The compound specific ranges of application
are listed in Table 7.3 and Table 7.4 together with the average ECso values.
The tested concentrations of chromium were not inhibiting at levels necessary to
calculate ECso values after 15 minutes.
7.2.3 Precision
The precision in terms of repeatability is presented in Table 7.5 and Table 7.6. The
repeatability is calculated for the six target compounds based on the results from Test
A.
Generally it was noticed that the repeatability was improved for ECso values compared
to EC20 values. For example, 30 minute LUMIStox £€20 values have RSDs for Zn2+,
C^O? " etc. as follows: 12, 55, 13, 73, 44 and 6.3. £€20 results are provided in the test
report /30/.
The log-log linearity, used by the model for EC calculation, was relatively low for
cyanide, causing high relative standard deviations.
18
-------
The precision in terms of reproducibility is presented in Table 7.7. Reproducibility is
based on the results from Test D, which was performed with Zn2+ as the target
compound. ECso values are closely related to the activity of the bacteria, as explained
in Section 7.2.4, 8.3.3, and in further details in the test report /30/.
Table 7.5 LUMIStox repeatability as relative standard deviation (RSD) in percent. For target
compounds in 2% NaCI MilliQ water. Number of replicates (n) is 3 but 4 for cyanide.
LUMIStox
Zn2+
Cr2072-
Triclosan
Cyanide
SDS
CTAB
15 min
EC50
RSD
(%)
4.5
n.a.
7.4
18
29
3.6
30 min
EC50
RSD
(%)
5.0
29
5.5
24
33
2.4
n.a.: Not applicable. EC50 could not be determined.
Table 7.6 ECLOX repeatability as relative standard deviation (RSD) in percent. For target compounds
in 2% NaCI MilliQ water. Number of replicates (n) is 3 but 4 for cyanide.
ECLOX
Zn2+
Cr2072-
Triclosan
Cyanide
SDS
CTAB
15 min
EC50
RSD
(%)
2.7
n.a.
4.6
15
34
6.3
30 min
EC50
RSD
(%)
4.9
24
2.2
16
38
1.2
n.a.: Not applicable. EC50 could not be determined.
Table 7.7 LUMIStox and ECLOX reproducibility as relative standard deviation (RSD) in percent. For
Zn2+ in 2% NaCI MilliQ water. Test was performed on three bacteria batches on three
different days. Number of replicates (n) is 3, except for ECLOX batch 02099 where 4
replicates were tested.
Znz+
LUMIStox
ECLOX
15 min
EC50
RSD
28
63
30 min
EC50
RSD
30
51
7.2.4 Agreement with accepted values
The agreement with accepted values was calculated for each target compound and the
average agreement was also determined (from Test A). The sources of accepted
literature values obtained with the ISO 11348-3 are listed in Appendix 3. The average
agreement was determined for all compounds which had literature values and where it
is known that the test was performed according to the ISO 11348-3.
19
-------
Table 7.8 LUMIStox £C5o agreement with accepted values (A) in percent.
Compound
Znz+
(ZnS04-7H20)
Cr2072-
(K2Cr207)
Triclosan
CTAB
Accepted values
EC50 ± 1 RSD
(mg/L)
2.2 ± 23%
19 ±11%
0.28
0.28
0.97
Test time
(min)
30
30
15
30
30
According to
ISO 11 348-3
Yes
Yes
Yes
Yes
Yes
LUMIStox
EC50±1 RSD
(mg/L)
4.1 ±4.4%
17 ±27%
0.40 ± 6.3%
0.53 ± 4.7%
0.97 ± 2.2%
A,
(%)
186
91
143
189
100
Table 7.9 ECLOX £C5o agreement with accepted values (A) in percent.
Compound
Znz+
(ZnSO4-7H2O)
Cr207z-
(K2Cr207)
Triclosan
CTAB
Acce
EC50 ± 1 RSD
(mg/L)
2.2 ± 23%
19 ±11%
0.28
0.28
0.97
pted values
Test time
(min)
30
30
15
30
30
According to
ISO 11 348-3
Yes
Yes
Yes
Yes
Yes
ECLOX
EC50 ± 1 RSD
(mg/L)
4.1 ±4. 3%
18±22%
0.39± 3.8%
0.53±2.3%
0.96± 1 .0%
A,
(%)
186
96
139
190
99
7.2.5
When evaluating the agreement with accepted values, it should be taken into account
that bacterial activity for some compounds affects the ECso values. It has been shown
that a low bacterial sensitivity, indicated by a low inhibition by the Zn2+ standard,
results in a higher ECso. For Test A, the activity of the bacteria caused an inhibition of
approximately 25% for the Zn2+ standard in a concentration that should equal ECso
according to the ISO 11348-3 method. The inhibition was therefore half of what could
be expected from the ECso value, but still within the accepted range from 20%-80%
inhibition, the acceptable range in the ISO 11348-3 method. The concentration needed
in Test A to obtain 50% inhibition was, due to the low bacteria activity, a factor of two
higher than the ECso value listed in the ISO 11348-3, and resulted in an agreement with
accepted value (Azn +) of 186%. Further details on this can be seen in the test report
/30/. The result for zinc is therefore seen as in general agreement with accepted values,
since the difference is explained by the bacteria activity, and the bacteria activity met
the requirements of the ISO 11348-3 method.
Robustness
Initial concentration, temperature, pH, color, turbidity and type of cuvettes
The robustness of the LUMIStox and ECLOX measurements was tested against
differences in initial concentration, temperature, pH, color, turbidity and type of
cuvettes. The robustness was calculated as the average inhibition under conditions of
the robustness test divided by average inhibition under reference conditions, and
reported as a percent.
20
-------
The results of the robustness test are both EC values (initial concentration) and percent
inhibition (all other robustness tests). The robustness under different test conditions is
listed in Table 7.10 to Table 7.13. Three different concentrations of dye were used for
color tests and three concentrations of BaSC>4 were used for turbidity tests.
Difference in initial concentration, temperature in laboratory, pH within requirements
listed in ISO 11348-3, and the use of plastic cuvettes in the LUMIStox caused
insignificant effects. The measurements and results show good robustness of the
methods and equipment for these parameters.
The use of ECLOX under field temperatures (5 °C and 23 °C) gave significantly
different results from the reference test conducted at 16 °C. The bacterial activity at 5
°C was generally low, resulting in high variation in the results. The robustness for the
two tested target compounds differed, showing that the robustness against field
temperature is compound specific.
Table 7.10 LUMIStox robustness (R) in percent. Test results are presented as EC values. R values
significantly different (95% confidence level, two-sided t-test) from 100% indicated in bold.
LUMIStox
Initial concentration
Ref. ~EC90
Target
compound
SDS
Condition
Initial concentration
~EC60
15 min
EC50
R
(%)
93
30 min
EC50
R
(%)
96
21
-------
Table 7.11 LUMIStox robustness (R) in percent. Test results are presented as % inhibition. R values
significantly different (95% confidence level, two-sided t-test) from 100% indicated in bold.
LUMIStox
Temperature, lab
Ref.15.4°C
PH
Ref. 7.0
Color
Ref. no color
Turbidity
Ref. no turbidity
Cuvette material1
Ref. glass
Target
compound
SDS
SDS
SDS
SDS
Zn2+
SDS
Condition
14.0 °C
16.1 °C
6.0
8.5
0.2% dye,
with c.c.
0.2% dye,
without c.c.
6.25% dye,
with c.c.
6.25% dye,
without c.c.
12. 5% dye,
with c.c.
12. 5% dye,
without c.c.
0.05 g BaS04/L,
with c.c.
0.05 g BaS04/L,
without c.c.
0.10gBaSO4/L,
with c.c.
0.10gBaSO4/L,
without c.c.
0.20 g BaSO4/L,
with c.c.
0.20 g BaSO4/L,
without c.c.
Plastic
Plastic
15 min
R
(%)
99
69
96
101
94
98
108
170
117
220
55
112
8
105
-90^
97
101 (99-160)
108(93-108)
30 min
R
(%)
105
71
110
107
102
105
107
156
114
197
70
106
41
97
-20^
88
107(106-
117)
99(90-101)
c.c.: Color correction.
1 Test performed in triplicates (with 3 replicates in each test). Median and interval are given as result.
2 Negative values occur when there inhibition is negative. Negative inhibition means that the solution tested gives
better growth conditions for the bacteria than the control.
22
-------
Table 7.12 ECLOX robustness (R) in percent. Test results are presented as EC values. R values
significantly different (95% confidence level, two-sided t-test) from 100% indicated in bold.
ECLOX
Initial concentration
Ref. ~EC90
Target
compound
SDS
Condition
Initial concentration
~EC60
15 min
EC50
R (%)
94
30 min
EC50
R (%)
97
n.a.: Not applicable.
Table 7.13 ECLOX robustness (R) in percent. Test results are presented as % inhibition. R values
significantly different (95% confidence level, two-sided t-test) from 100% indicated in bold.
Number of replicates are 3, except for Test I on turbidity where the number of replicates is
4.
ECLOX
Temperature, field1
Ref. 16°C
Temperature, lab
Ref.15.4°C
PH
Ref. 7.0
Color
Ref. no color
Turbidity
Ref. no turbidity
Target
compound
Znz+
SDS
SDS
SDS
SDS
SDS
Condition
5°C
23 °C
5°C
23 °C
14.0 °C
16.1 °C
6.0
8.5
0.2% dye,
with c.c.
0.2% dye,
without c.c.
6.25% dye,
with c.c.
6.25% dye,
without c.c.
12. 5% dye,
with c.c.
12. 5% dye,
without c.c.
0.05 g BaSO4/L,
with c.c.
0.05 g BaSO4/L,
without c.c.
0.10gBaSO4/L,
with c.c.
0.10gBaSO4/L,
without c.c.
0.20 g BaSO4/L,
with c.c.
0.20 g BaSO4/L,
without c.c.
15 min
R
(%)
27(11-35)
116(108-171)
100(93-105)
75 (73-75)
88
91
111
101
124
105
107
155
128
214
135
109
154
118
115
92
30 min
R
(%)
n.d.
n.d.
n.d.
n.d.
100
85
113
105
124
110
112
148
115
180
111
93
130
107
101
86
n.d.: Not determined.
c.c.: Color correction.
1 Performed at three different concentrations.
Median and interval are given as result.
23
-------
The results showed that the use of color correction is essential when testing colored
samples, while the results for turbid BaSC>4 samples showed that the color correction
function is not applicable. However, this could differ for other types of turbid samples.
Additional testing is needed for verification.
According to ISO 11348-3 strongly turbid samples should be allowed to settle for 1 h,
centrifuged, or be filtered. The robustness test performed was done on turbid samples
(no settling, centrifugation, or filtration) to see the effect if these guidelines were not
followed. The HACH-LANGE manuals are not clear in this recommendation from the
ISO standard. It is suggested that the vendor revise the manuals to ensure the best test
results for turbid samples.
Wastewater matrix
Wastewater contains ions, organic compounds and particles which may potentially
alter the detected toxicity of substances in the wastewater by processes such as
complexation and adsorption. Two non-toxic wastewater types (industrial and
domestic) were therefore used as the matrix and compared to 2% NaCl MilliQ water.
The wastewaters were evaluated for toxicity as part of Test K using the dilution series
to determine EC20 and ECso. £€20 or ECso values generated were not calculated when
the inhibition was lower than 10%. Originally, triplicate measurements were to be
made; however, during testing an error resulted in the loss of one replicate for
industrial wastewater. It was decided that two replicate measurements were sufficient
to document that both the domestic wastewater and the domestic wastewater were non-
toxic. This is documented in Deviations 5 and 6 (see Appendix 4). Individual inhibition
measurements of the wastewaters that were based on triplicate measurements as part of
Test J are included in Table 8.14 below.
The baseline luminescence of the non-toxic wastewater differed slightly from the
baseline of the 2% NaCl MilliQ water, illustrated in Table 7.14. The domestic
wastewater appears to enhance the luminescence, causing negative inhibition.
Table 7.14 Wastewater baseline luminescence given as % inhibition. Number of replicates is 3.
Wastewater
Industrial
Domestic
LUMIStox
15 min
% inhibition
1.2
-8.1
30 min
% inhibition
1.5
-5.7
ECLOX
15 min
% inhibition
-2.9
-6.6
30 min
% inhibition
-3.3
-5.3
Table 7.15 and Table 7.16 show the results of robustness towards wastewater. The
domestic wastewater is reported both with and without an adjustment to the baseline to
account for the wastewater's negative inhibition (positive growth effect) on the
bacterial luminescence (see Table 7.14).
24
-------
Table 7.15 LUMIStox robustness (R) towards wastewater given in percent. R values significantly
different (95% confidence level, two-sided t-test) from 100% indicated in bold.
LUMIStox
Matrix
Ref. 2% NaCI
MilliQ water
Target
compound and
concentration
Zn2+
4.0 mg/L
Cr2072-
2.8 mg/L
Triclosan
0.60 mg/L
SDS
0.80 mg/L
CTAB
1 .2 mg/L
Wastewater
Industrial
Domestic
Industrial
Domestic
Industrial
Domestic
Industrial
Domestic
Industrial
Domestic
15 min
Inhibition
R
(%)
77
31
31
-501
114
84
68
66
102
75
30 min
Inhibition
R
(%)
43
84
0
-10 1
141
57
28
64
68
52
Adjusted baseline
15 min
Inhibition
R
(%)
127
15
105
107
118
30 min
Inhibition
R
(%)
123
22
96
96
78
1 Negative value occurs when inhibition is negative. Negative inhibition means that the solution tested gives better
growth conditions for the bacteria than the control.
Table 7.16 ECLOX robustness (R) towards wastewater given in percent. R values significantly
different (95% confidence level, two-sided t-test) from 100% indicated in bold.
ECLOX
Matrix
Ref. 2% NaCI
MilliQ water
Target
compound
Zn2+
4.0 mg/L
Cr2072-
2.8 mg/L
Triclosan
0.60 mg/L
SDS
0.80 mg/L
CTAB
1 .2 mg/L
Wastewater
Industrial
Domestic
Industrial
Domestic
Industrial
Domestic
Industrial
Domestic
Industrial
Domestic
15 min
Inhibition
R
(%)
56
37
12
-60 1
116
89
68
71
99
64
30 min
Inhibition
R
(%)
22
85
-20 1
-10 1
141
62
35
67
61
49
Adjusted baseline
15 min
Inhibition
R
(%)
132
14
110
111
101
30 min
Inhibition
R
(%)
125
13
101
101
73
1 Negative value occurs when inhibition is negative. Negative inhibition means that the solution tested gives better
growth conditions for the bacteria than the control.
Chromium showed a change in toxicity when added to the wastewater, but effects are
also seen in some cases for zinc, SDS and CTAB.
7.3 Evaluation of test data quality
7.3.1 Reference chemical analysis performance data
Control data for the reference chemical analysis obtained from Eurofms are
summarized in Table 7.17.
25
-------
Table 7.17 Performance parameters for reference chemical analysis control data.
Target compound
Znz+
Cr2072-
Triclosan
Cyanide (CN~)
SDS (anionic surfactants1)
CTAB (cationic surfactants2)
Limit of detection
ug/L
0.50
0.50
0.10
1.0
25
100
Precision (RSD)
%
15
15
Not specified
10
15
20
Trueness
%
98-99
103
103
99
101
95
1 Reference compound is SDS.
2 Reference compound is benzyl di-methyl tetradecyl ammonium chloride-dihydrate, molar weight 404,00 g/mol.
Table 7.19, in section 7.3.3, lists analyses of blank samples, performed to test the
Eurofins detection limits. Eurofms participates in proficiency tests for most of the
tested compounds. The results of their most recent proficiency tests are shown in Table
7.18.
Table 7.18 Results of Eurofins proficiency tests.
Parameter
Zinc
Chromium
Triclosan
Cyanide
Anionic
surfactants
Cationic
surfactants
Nominal
value
61 4 ug/L
83.1 ug/L
Zeta-score
0.316
0.157
Supplier
APG, November 2009 WS, 1. round
FAPAS (LEAP), Wastewater, G20+G21
Eurofins has not participated in proficiency testing, since triclosan is a
new parameter for them and is not covered by their accreditation
7.00-1 1.3 ug/L
50.0-1 20 ug/L
0.377
-0.464
KIWA, drinking water, 09-03
KIWA, drinking water, 09-03
Eurofins is not aware of supplier of proficiency tests for cationic
surfactants within the measuring area
7.3.2
7.3.3
Comparable test performance data
ALcontrol uses zinc sulfate and phenol as reference compounds. The results of the data
were within the specification of the bacteria supplier, though the control chart for zinc
shows that over the period the references have been at a low level, around 70% of the
expected average.
ALcontrol participates in an annual proficiency test with the Microtox®. The results
were audited by Battelle as part of the technical systems audit (TSA) at ALcontrol and
found to be within the acceptance criteria.
Test system control data
Blank samples
The 2% NaCl MilliQ water used to prepare stock solutions of test compounds was
tested for background levels of the target compounds. The results are shown in Table
7.19.
The results showed that the 2% NaCl MilliQ water did not contain any of the target
compounds in significant concentrations. These results also showed that the water
purifier was operating within normal parameters, and the NaCl was free of
contaminants.
26
-------
Table 7.19 Concentrations of target compounds in 2% NaCI MilliQ water (blank) samples.
Target compound
Zn2+
Cr2072-
Triclosan
Cyanide (CN~)
SDS (anionic surfactants1)
CTAB (cationic surfactants^)
Concentration
ug/L
Replicate 1
<0.50
0.50
<0.10
<1.0
<25
<100
Replicate 2
<0.50
0.60
0.19
<1.0
<25
<100
1 Reference compound for anionic surfactant is SDS.
2 Reference compound for cationic surfactant is benzyl di-methyl tetradecyl ammonium chloride-dihydrate.
The 2% NaCI MilliQ water was tested for toxicity at ALcontrol.
The results are shown in Table 7.20.
Table 7.20 Toxicity in percentage of sample volume of 2% NaCI MilliQ water (blank) samples.
Time
(min)
5
15
30
EC value
EC20
EC50
EC20
EC50
EC20
EC50
Concentration
%
78
>82
>82
>82
>82
>82
The results showed no detectable toxicity of the 2% NaCI MilliQ water after 15 and 30
minutes.
Control, stock solutions
The concentrations and the stability of the stock solutions were evaluated by sending
subsamples of the solutions to Eurofms laboratory for chemical analysis. Table 7.21
shows the results of this analysis and the recovery of the concentrations in the stock
solutions.
The surfactants SDS and CTAB were expected to adhere to the cuvettes. In addition,
CTAB was difficult to dissolve. The stock solutions were therefore treated as the test
samples (added to cuvettes and left for 30 minutes) before sending to Eurofms.
27
-------
Table 7.21 Concentrations (average and relevant range (high/low value divided by average)) of target
compounds in spiked 2% NaCI MHHQ water stock solutions.
Target compound
Zn2+
Cr2072-
Triclosan
Cyanide (CN~)
SDS (anionic surfactants1)
CTAB (cationic surfactants^)
CTAB3
Measured concentration
Average
UQ/L
17,500
52,000
355
31,500
2,550
725
560
Relevant range
%
±5.7
±7.7
±2.8
±9.5
±12
±32
±32
Prepared
concentration
ug/L
22,000
56,100
1,600
32,885
35,950
30,000
Recovery
%
80
93
22
96
7.1
1.9
1 Reference compound is SDS.
2 Reference compound is benzyl di-methyl tetradecyl ammonium chloride-dihydrate ((CeHsC
C14Alkyl)+Cr), molar weight 404.00 g/mol. Molar weight of reference compound cation 368.5 g/mol.
3 Concentration of CTAB, molar weight 364.45 g/mol has been calculated based on CTAB (cationic surfactants)
results. Molar weight of CTAB cation 284.5 g/mol.
The concentration of SDS and CTAB in the cuvettes were found to be low compared to
the expected (7% and 2% recovery). The triclosan stock solution also showed a
significant loss, with a recovery of only 22%. These losses were not taken into account
when calculating the EC values. This was noted in Table 7.3 and Table 7.4. Despite the
low recoveries of triclosan and CTAB we found ECso values similar to the accepted
literature values (Table 7.8 and Table 7.9). The reason for the low recovery has not
been determined in this verification. Therefore, a conclusion regarding the risk of
reduced inhibition due to losses during handling of the samples cannot be drawn from
these results. It is noted that chemical analysis of the sample is not required in the ISO
11348-3.
The concentration of cyanide in the dilutions was determined using a test kit. An
artificial cyanide sample was carried through the test procedure. Instead of adding
bacteria solution, 2% NaCI was added. No measurements of luminescence were
performed. Instead, the cyanide concentration was measured using a HACH-LANGE
test (LCK 315). Test row B was analyzed at time 0 and test row C was analyzed after
30 minutes. The results showed that cyanide was stable during the test, i.e. the
concentration after 30 minutes was within the acceptable range of 80-120% of the
initial concentrations. Cyanide was therefore included in the test program.
Test of inhibition by reference substances according to ISO 11348-3
The bacterial batches used in the tests were tested for compliance with the
requirements in the ISO 11348-3, section 11. For all reference standard compounds the
criteria is 20%-80% inhibition. The results of the reference tests are shown in Table
7.22.
28
-------
Table 7.22 Mean %-inhibition and standard deviation (st. dev.) from reference tests of bacteria batches
performed in accordance with ISO 11348-3. Tests are performed on LUMIStox, number of
replicates are two except for 02099 where the reference standards were only tested once.
Batch
10129
11169
02099
ISO requirement
Zn2+
(2.2 mg/L)
%
22 ±0.03
36 ±0.90
15
Cr20/-
(18.7 mg/L)
%
60 ±0.12
53 ±1.4
96
3,5-dichlorophenol
(3.4 mg/L)
%
20 ±2. 3
28 ± 0.62
39
20-80
f)\
For batches 10129 and 11169, both Zn and 3,5-dichlorophenol were close to the
lower limit of 20%. One tested sample was below 20% for 3,5-dichlorophenol. The
reference standard, Zn2+ was included in all tests to be able to follow the bacteria
activity.
The bacteria batch 02099 was only used in one test (Test D). The results for this
bacteria batch did not meet the replicates required on the percent inhibition to fulfill
the requirement of the ISO standard. Use of this bacteria batch is documented in
Deviation 7 (see Appendix 4). This resulted in slightly higher standard deviations and
higher relative standard deviations reported in Table 8.7. As noted in Deviation 3, this
batch was used even though it did not fulfill the ISO requirement, because at least three
batches were needed for the reproducibility evaluation and no other bacteria batches
were immediately available from the vendor.
2+N
Since the references were close to the requirements in the ISO, one reference (Zn )
was included in all test runs. This is more stringent than what is stated in the ISO
standard. Few of the Zn-reference standard test results (approximately 10% of those
measured in the LUMIStox) did not fulfill the ISO requirement. However, all results
have been included in the evaluation since the check with the original reconstitution
(reported in Table 7.22) fulfilled the ISO requirements.
The ISO standard also sets limits for the variation (i.e., deviation) between the
duplicate control sample measurements. Duplicate measurements should not deviate
from their mean by more than 3%. HACH-LANGE informed DHI that for the ECLOX,
this can be difficult to fulfill. In some cases the deviation of ECso values between
triplicate samples was as low as 1.2%, even though the deviation between the replicate
control measurements were above the required 3%. No data has therefore been
excluded because of the deviation between duplicate control measurements.
Test of samples by the external laboratory
A solution of 96.7 mg/L of zinc sulfate heptahydrate (22 mg/L Zn2+) in 2% NaCl
MilliQ water was tested for toxicity at ALcontrol. The results are shown in Table 7.23.
ECso,3o mm was considerably lower than the accepted value (see Table 7.8).
29
-------
Table 7.23 Toxicity in percentage of sample volume of 96.7 mg/L zinc sulfate heptahydrate (22 mg/L
Zn2+) in 2% NaCI MilliQ water.
Time
(min)
5
15
30
EC value
EC20
EC50
EC20
EC50
EC20
EC50
Concentration
%
19
>82
2.5
12
0.20
1
"Concentration
(mg Zn2+/L)
4.2
>18
0.60
2.6
0.05
0.20
* Calculated nominal concentration based on added amount. See also Table 7.21.
A solution containing 7.2 mg/L SDS in 2% NaCI MilliQ water was tested for toxicity
at ALcontrol. The results are shown in Table 7.24. These results are similar to the
results obtained with the both LUMIStox (Table 7.3) and ECLOX (Table 7.4).
Table 7.24 Toxicity in percentage of sample volume of 2% NaCI MilliQ water added 7.2 mg/l of SDS.
Time
(min)
5
15
30
EC value
EC20
EC50
EC20
EC50
EC20
EC50
Concentration
%
9
18
6
11
4
9
"Concentration
(mg SDS/L)
0.60
1.3
0.40
0.90
0.30
0.60
"Calculated nominal concentration based on added amount. There was a low recovery of SDS as
determined by the chemical analyses. See Table 7.21.
7.3.4 Audits
Two onsite audits were performed during the testing. An internal audit performed by
Bodil Mose Pedersen from DHI /31/ resulted in one deviation in the internal protocol
(Appendix 4 in the test plan 1251). The deviation was resolved directly after the audit.
An on-site audit was performed by Battelle AMS Center for U.S. EPA /32/. One
finding, four observations and one recommendation were noted. The final audit report
is permanently stored with the Battelle AMS Center Quality Manager.
7.3.5 Deviations
There were no amendments to the verification protocol or the test plan.
Four deviations were made to the verification protocol.
There have been 10 deviations to the test plan, all deviations have been approved. The
test report reflects these deviations.
All deviations to the verification protocol and test plan are included in Appendix 4.
7.4 Additional parameters summary
7.4.1 User manual
The assessment for the user manual evaluated if the manual describes the use of the
equipment adequately. The evaluation considered whether the manual was
30
-------
understandable for a typical laboratory technician. This evaluation was based on a
number of specific points of importance; see Table 7.25 and Table 7.26 for the
parameters included and the assessment outcomes.
Table 7.25 Assessment of the user manual for LUMIStox.
Parameter
Product
Principle of
operation
Intended use
Performance
expected
Limitations
Preparations
Unpacking
Transport
Assembly
Installation
Function test
Operation
Steps of operation
Points of caution
Accessories
Maintenance
Trouble shooting
Safety
Chemicals
Power
Complete
description
V
V
V
V
V
V
V
V
V
V
V
V
Summary
description
V
No
description
V
Not
relevant
V
V
31
-------
Table 7.26 Assessment of the user manual for ECLOX.
Parameter
Product
Principle of
operation
Intended use
Performance
expected
Limitations
Preparations
Unpacking
Transport
Assembly
Installation
Function test
Operation
Steps of operation
Points of caution
Accessories
Maintenance
Trouble shooting
Safety
Chemicals
Power
Complete
description
V
V
V
V
V
V
V
V
V
V
V
V
V
V
Summary
description
V
No
description
V
Not
relevant
A description was considered complete if all essential steps were described, if they
were illustrated with a figure or a photo, where relevant, and if the descriptions were
understandable without reference to other guidance.
7.4.2 Product costs
The capital investment costs and the operation and maintenance cost - components of
product sustainability - were itemized based upon a determined design basis 7287; see
Table 7.27 for the items that were included.
The design basis was determined based on one laboratory day. According to HACH-
LANGE the shelf life of the dried reagent is one year, the lifetime of the rehydrated
bacteria suspension is 4 hours. Within that time it was possible to perform an ECso test
according to the ISO 11348-3 on three samples plus associated controls and standards.
32
-------
Table 7.27 List of capital cost items and operation and maintenance cost items per product unit.
Item type
Capital
Buildings and land
Equipment
Utility connections
Installation
Start up/training
Operation and maintenance
Materials,
including chemicals
Utilities,
including water and energy
Labor
Item
Laboratory facility
LUMIStoxorECLOX
LUMIStherm
PC with LUMISsoft4
Power supply
Can be done by
operator/laboratory
technician
Training of laboratory
technician
Bacteria
Cuvettes
Reconstitution solution
2% NaCI solution
Solid NaCI
Power
One laboratory technician
Number/duration
1
1
1
1
3
1 day
1 day
1 batch
20 pr. sample (=60 per day)
1 bottle
1 bottle
1 bottle
PC and screen ~6 kWh
ECLOX 4 AA batteries
LUMIStox -0.4 kWh
LUMIStherm -0.4 kWh
1 day
Costs associated with the equipment at the time of testing were:
• LUMIStox, LUMIStherm, the software LUMISoft4: 13,000 Euro (17,800 $U.S.),
• ECLOX, LUMIStherm, the software LUMISoft4: 6,500 Euro (9,600 $U.S.).
Additional equipment such as cuvettes, bacteria and chemicals on a cost-per-sample
basis as used for testing for EC50 according to the ISO 11348-3: 18 Euro (23 $U.S.).
7.4.3 Occupational health and environment
The risks for occupational health and safety and for the environment associated with
the use of the products were compiled. The compilation emphasized chemicals
classified as hazardous used during product operation 7297. No application of
hazardous chemicals was identified during testing.
No risk from installation, operating and maintaining the product were identified, based
on an assessment of risks for human health, power supply, and danger of infections. No
additional risks compared to conventional effluent wastewater testing or analyses were
identified.
7.5 Operational parameters
The effluent wastewater parameters covered in the test are summarized in Table 7.28.
The wastewater parameters were measured by Eurofins.
33
-------
Table 7.28 Results of analytical parameters analyzed in wastewater.
Parameters
Turbidity
TOC
Conductivity
Alkalinity
PH
COD
Suspended solids (SS)
Nitrogen (total)
Phosphorus (total)
BOD5
Unit
FTU
mg/L
mS/m
mmol/L
-
mg/L
mg/L
mg/L
mg/L
mg/L
Industrial wastewater
15
39
4300
6.9
7.7
110
83
6.3
4.2
3.4
Domestic wastewater
2.4
10
140
5.5
7.5
28
4.9
6.9
0.23
5.2
The operational parameters tested in the DHI laboratory are summarized in Table 7.29.
Table 7.29 Operational parameters evaluated during testing.
Temperature of
thermal block
14.0- 16.1 °C
pH in sample
6.0-8.5
Color
correction
Colored
samples Turbid
samples
Temperature at
field use
(ECLOX)
5 - 23 °C
Cuvette
material
(LUMIStox)
Glass
Plastic
7.6 Recommendation for verification statement
The verification statement is a summary of the results described in the verification
report. The results included in the verification statement are listed in this section.
Table 7.30 Description of matrix and effect for LUMIStox and ECLOX.
Matrix
Effect
LUMIStox and ECLOX are applied for
wastewater; river and lake water; leachate
from soil, waste, rubble, etc.; or directly in
fluent chemicals. Verification testing was
conducted on domestic and industrial
wastewater effluents
Measurement of toxicity as indicated by
inhibition of luminescent bacteria by a
variety of compounds including metal ions,
organic pesticides, inorganic and organic
pollutants and surfactants
Additional parameters included:
User manual quality, product cost,
environmental health and safety
The primary results include short description of the matrix and effect as given in Table
7.30, the performance parameters verified for LUMIStox and given in Table 7.31 and
for ECLOX, given in Table 7.32. Listed are results for ECso values or percent
inhibition after 30 minutes.
34
-------
Table 7.31
LUMIStox performance parameter summary.
LUMIStox
Criterion of
detection
Range of
application
Precision
Agree-
ment with
accepted
values
Robustness
Compound
Repeat-
ability
% inhibition
Reproduc-
ibility
pH, color,
turbidity,
laboratory
temperature11
Cuvette
material
Waste-
water
matrix11
General
Zn2
Cr2Cy2-
Triclosan
anide
SDS
CTAB
' For colored samples are given robustness after use of color correction. For BaSO4-turbide samples is given
robustness without use of color correction. For domestic wastewater adjustment was made to account for the negative
inhibition from the wastewater, if color correction was used the robustness was -20% to 70%. The values given are
therefore the best achievable robustness.
Table 7.32 ECLOX performance parameter summary.
ECLOX
Criterion of
detection
Range of
application
Precision
Agree-
ment with
accepted
values
Robustness
Compound
Repeat-
ability
% inhibition
Reproduc-
ibility
pH, color,
turbidity,
laboratory
temperature11
Field
temperature
(15 minutes)
Waste-
water
matrix11
Cr2Cy2-
Triclosan
anide
SDS
CTAB
For colored samples are given robustness after use of color correction. For BaSO4-turbide samples is given
robustness without use of color correction. For domestic wastewater adjustment was made to account for the
negative inhibition from the wastewater, if color correction was used the robustness was 101 % to 130%. The
values given are therefore the best achievable robustness.
The user manual and other instructions were described as complete. The manual
described that color correction shall be used for colored as well as turbid samples. The
robustness test with BaSO4-turbid samples showed that application of color correction
was not appropriate.
The product costs based on a scenario for one laboratory day are as listed:
35
-------
Table 7.33 List of capital cost items and operation and maintenance cost items per product unit.
Item type
Capital
Buildings and land
Equipment
Utility connections
Installation
Start up/training
Operation and maintenance
Materials, including
chemicals
Utilities, including water and
energy
Labor
Item
Laboratory facility
LUMIStoxorECLOX
LUMIStherm
PC with LUMISsoft4
Power supply
Can be done by
operator/laboratory
technician
Training of laboratory
technician
Bacteria
Cuvettes
Reconstitution solution
2% NaCI solution
Solid NaCI
Power
One laboratory technician
Number/duration
1
1
1
1
3
1 day
1 day
1 batch
20 pr. sample (=60 per day)
1 bottle
1 bottle
1 bottle
PC and screen ~6 kWh
ECLOX 4 AA batteries
LUMIStox -0.4 kWh
LUMIStherm -0.4 kWh
1 day
Costs associated with the equipment at the time of testing were:
• LUMIStox, LUMIStherm, the software LUMISoft4: 13,000 Euro (17,800 $U.S.),
• ECLOX, LUMIStherm, the software LUMISoft4: 6,500 Euro (9,600 $U.S.).
Additional equipment such as cuvettes, bacteria and chemicals on a cost-per-sample
basis as used for testing for EC50 according to the ISO 11348-3: 18 Euro (23 $U.S.).
Application of the test systems does not give rise to any special risk or contact to
hazardous substances other than what occur doing conventional testing of wastewater
effluents.
The operational parameters are shown in Table 7.34 and the wastewater chemistry is
listed in Table 7.35 as range of concentration or parameter measured.
Table 7.34 Operational parameters evaluated during testing.
Temperature of
thermal block
(°C)
14.0-16.1
pH in sample
6.0-8.5
Color
correction
Colored samples
Turbid samples
Temperature at
field use
(ECLOX)
5 - 23 °C
Cuvette
material
(LUMIStox)
Glass
Plastic
36
-------
Table 7.35 Range of analytical parameters analyzed in wastewater.
Parameters
Turbidity
TOC
Conductivity
Alkalinity
PH
COD
Suspended solids (SS)
Nitrogen (total)
Phosphorus (total)
BOD5
Unit
FTU
mg/L
mS/m
mmol/L
-
mg/L
mg/L
mg/L
mg/L
mg/L
Range
2.4-15
10-39
140-4300
5.5-6.9
7.5-7.7
28- 110
4.9-83
6.3-6.9
0.23-4.2
3.4-5.2
37
-------
8 VERIFICA TION SCHEDULE
The verification was planned and performed from October 2009 through April 2010.
The overall schedule is provided in Table 8.1.
Table 8.1
Verification schedule.
Task
Quick scan
Verification protocol and test plan
Test
Test reporting
Verification
Verification report
Report and verification statement
preparation and review
Timing
October 2009
October to December 2009
January to April 2010
February to April 2010
April 2010
April 2010
April 2010 to February 201 1
38
-------
QUALITY ASSURANCE
The quality assurance of the verification is described in Table 9.1 and Fig 1.2. The
quality assurance of the tests was described in the test plan as well as in the process
document prepared by Battelle /!/, and is summarized in Table 10.1 below.
Table 9.1 QA summary for the verification.
Initials
Tasks
Plan document with
verification protocol
and test plan
Test system
Test report
Verification report
DHI
MWN
Re vie
w
_
Re vie
w
Re vie
w
BOP
Audit
-
-
Battelle
AMS
Center
ZW
Review
Audit
-
Audit/Revie
w
U.S. EPA
ETV
JMK, MH
Review
_
-
Review
ETV
Canada
MEH
Review
_
Review
Review
Environ-
ment
Canada
BD
_
-
Review
Expert Group
KOK, JA, AA,
ML
(verification
report only)
Review
_
Review by
KOK
Review
An internal review of plan and report documents was conducted by the Head of
Innovation, Margrethe Winther-Nielsen (MWN). A test system audit (see test plan)
was conducted on 22 January 2010 following GLP audit procedures by a trained
auditor: Senior Chemical Engineer, Bodil Mose Pedersen (BOP).
The Battelle Quality Manager, Zachary Willenberg (ZW) performed a technical
systems audit (TSA) during this verification and test on 26-29 January 2010. An audit
of data quality was conducted 13 May through 4 June 2010.
U.S. EPA staff, John McKernan (JMK) and Michelle Henderson (MH) and Mona El-
Hallak (MEH) from ETV Canada reviewed all plan and report documents, except the
test report. In addition, Mona El-Hallak (MEH) from ETV Canada reviewed the test
report, and Benoit Desforges (BD) from Environment Canada reviewed the verification
report.
The expert group, Kresten Ole Kusk (KOK), Dr. Joel Allen (JA) and Dr. Ali Amiri
(AA) reviewed the plan and report documents, though only Kresten Ole Kusk (KOK)
reviewed the test report. Dr. Max Lee also reviewed the verification report.
39
-------
APPENDIX 1
Terms and definitions used in the verification protocol
40
-------
The abbreviations and definitions used in the verification protocol are summarized
below.
Where discrepancies exist between DANETV and U.S. EPA ETV terminology,
definitions from both schemes are given.
Word
Agreement
with accepted
values
AMS Center
Analytical
laboratory
Application
CD
CTAB
DANETV ETV
EC
ECLOX
Effect
EN
ETV
EU
Evaluation
Experts
GLP
ISO
LID
Limit of
detection
LoD
LUMISsoft4
LUMIStherm
LUMIStox
Matrix
DANETV
Here defined as the % agreement between
literature values and test results
Advanced Monitoring Systems Center at
Battelle
Independent analytical laboratory used to
analyze reference samples
The use of a product specified with respect
to matrix, target, effect and limitations
Criterion of detection
Cetyl trimethyl ammonium bromide
The Danish Centre for Verification of
Climate and Environmental Technologies
Effect concentration, e.g. causing 50%
inhibition (EC50)
ECLOX handheld luminometerfrom
HACH-LANGE
The way the target is affected
European standard
Environmental technology verification
(ETV) is an independent (third party)
assessment of the performance of a
technology or a product for a specified
application, under defined conditions and
adequate quality assurance
European Union
Evaluation of test data for a technology
product for performance and data quality
Independent persons qualified on a
technology in verification or on verification
as a process
Good Laboratory Practice
International Standardization Organization
Lowest ineffective dilution. Often seen as
the dilution in a dilution series causing less
than 20% inhibition
Calculated from the standard deviation of
replicate measurements at less than 5
times the detection limit evaluated.
Corresponding to less than 5% risk of false
blanks
PC software from HACH-LANGE,
produced for LUMIStox
Thermostat from HACH-LANGE, produced
for LUMIStox
LUMIStox 300 bench top luminometer from
HACH-LANGE
The type of material that the product is
intended for
U.S. EPA ETV
EPA program that develops generic
verification protocols and verifies the
performance of innovative environmental
technologies that have the potential to
improve protection of human health and
the environment
An examination of the efficiency of a
technology
Peer reviewers selected for a verification
41
-------
Word
Method
OD
PC
Performance
claim
Performance
parameters
Precision
(Environmen-
tal) product
QA
Range of
application
Reference
analyses
Repeatability
Reproducibility
Robustness
RSD
SDS
Stakeholder
Standard
DANETV
Generic document that provides rules,
guidelines or characteristics for tests or
analysis
Optical density
Personal computer
The effects foreseen by the vendor on the
target (s) in the matrix of intended use
Parameters that can be documented
quantitatively in tests and that provide the
relevant information on the performance of
an environmental technology product
The relative standard deviation obtained
from replicate measurements, here
measured under repeatability or
reproducibility conditions
Ready to market or prototype stage
product, process, system or service based
upon an environmental technology
Quality assurance
Generally: the range from the LoD to the
highest concentration with linear response.
For this verification the range is based on
range of dilution of a test sample
Analysis by a specified reference method
in an accredited (ISO 17025) laboratory
The precision obtained under repeatability
conditions, that is with the same
measurement procedure, same operators,
same measuring system, same operating
conditions and same location, and
replicate measurements on the same or
similar objects over a short period of
time
The precision obtained under
reproducibility conditions. Measurement
performed at different locations, operators,
measuring systems, and replicate
measurements on the same or similar
objects
% variation in measurements resulting
from defined changes in matrix properties
Relative standard deviation in %
Sodium lauryl sulphate
Generic document established by
consensus and approved by a recognized
standardization body that provides rules,
guidelines or characteristics for tests or
U.S. EPA ETV
(Environmental) technology
Buyers and users of technology,
technology developers/vendors, the
consulting engineers, the finance and
export communities, government
permitters, regulators, first responders,
emergency response, disaster planners,
public interest groups, and other groups
interested in the performance of
innovative environmental technologies
42
-------
Word
Target
(Environmen-
tal) technology
Test/testing
Trueness
ISA
U.S. EPA
Vendor
Verification
Vibrio fischeri
DANETV
analysis
The measurable property that is affected
by the product
The practical application of knowledge in
the environmental area
Determination of the performance of a
product by parameters defined for the
application
The % recovery of true value obtained
either from knowledge on the preparation
of test solutions or from measurements
with reference methods
Technical system audit
United States Environmental Protection
Agency
The party delivering the product or service
to the customer
Evaluation of product performance
parameters for a specified application
under defined conditions and adequate
quality assurance
Light producing bacteria used in
luminescent bacteria test
U.S. EPA ETV
An all-inclusive term used to describe
pollution control devices, controls,
monitoring systems, waste treatment
processes and storage facilities, and site
remediation technologies and their
components that may be utilized to
remove pollutants or contaminants from,
or to prevent them from entering the
environment
The technology developer, owner, or
licensee seeking verification
Establishing or proving facts of the
performance of a technology under
specific, predetermined criteria, test
plans and adequate data QA procedures
43
-------
APPENDIX 2
References
44
-------
1. DANETV Test Environmental Technology Verification Program. Advanced Monitoring Systems Center.
Process Document for U.S. EPA ETV AMS Center, DHI DANETV Water Centre and ETV Canada. Joint
Verification of the HACH-LANGE GmbH LUMIStox 300 Bench Top Luminometer and ECLOX
Handheld Luminometer. Battelle. January 2010.
2. Battelle. Quality management plan (QMP) for the ETV Advanced Monitoring Systems Center. Version
7.0. Dated November 2008.
3. ETV Canada General Verification Protocol. February 2007.
4. DANETV Test Center and Test Organization. Center Quality Manual - Water technology. Version 2.
October 2009.
5. HACH-LANGE. ECLOX Tox LUMIStox specs and features. Data sheet. Received from HACH-LANGE
autumn 2009.
6. HACH-LANGE. LUMIStox workshop. Power point presentation. Received from HACH-LANGE autumn
2009.
7. ETV, EPA U.S. Environmental Protection Agency. Environmental Technology Verification Report.
Strategic Diagnostics Inc. Microtox. Rapid toxicity testing system. Battelle. November 2003.
8. ETV, EPA U.S. Environmental Protection Agency. Environmental Technology Verification Report.
Strategic Diagnostics Inc. Deltatox. Rapid toxicity testing system. Battelle. November 2003.
9. ETV, EPA U.S. Environmental Protection Agency. Environmental Technology Verification Report.
CheckLight, Ltd. ToxScreen-II. Rapid toxicity testing system. Battelle. November 2003.
10. TESTNET. D3.1a Evaluation report test case la: TOXcontrol BioMonitor for surface water. DHI. 01-03-
2008.
11. Machery-Nagel. Rapid Tests. Water Analysis. Page 128-130. Located on www.mn-net.com, 24-09-2009.
12. HIDEX. New compact three in one personal multilabel tester Triathler. 2002. Located on
www.hidex.com, 24-09-2009.
13. Dr Lange. Correcting absorptive inhibition in the luminescent bacteria test by means of a combined
luminometric/photometric procedure. INFO 13. January 1998.
14. Directive 2008/105/EC of the European parliament and the council on environmental quality standards in
the field of water policy. 16 December 2008.
15. Bekendtgorelse om miljokrav for vadomrader og krav til udledning af forurenende staffer til vandlob, soer
eller havet. Bekendtgorelse nr. 1669. Danish Ministry of the Environment. Dated 14. December 2006.
16. Miljoministeriet. By- og Landskabsstyrelsen. Oversigt over forslag til kvalitetskriterier for staffer omfattet
afbilag 1 i bekendtgorelse nr. 1669/2006 om miljokrav for vandomradermv. J. Nr. BLS-489-00051.
Dated 7. October 2009.
45
-------
17. ISO. General requirements for the competence of testing and calibration laboratories. ISO 17025. 2005.
18. International Standardization Organisation. EN ISO 9001. Quality management systems - Requirements.
15-11-2008.
19. Gavaskar, A. and Gumming, L. Cost Evaluation Strategies for Technologies Tested under the
Environmental Technology Verification Program. 2001. Battelle.
20. European Commission. Commission Directive on classification, packaging and labelling of dangerous
substances. 2001/59/EC. 2001.
21. Farre, M., Asperger, D. Kantiani, L., Gonzalez, S., Petrovic, M. & Barcelo, D. (2008) Assessment of the
acute toxicity of triclosan and methyl triclosan in wastewater based on the bioluminescence inhibition of
Vibrio fischeri. Anal Bioanal Chem 390:1999-2007.
22. Romanelli, M.F., Moraes, M.C.F., Villavicencio, A.L.C.H. andBorrely, S.I. (2004) Evaluation of toxicity
reduction of sodium dodecyl sulfate submitted to electron beam radiation. Radiation Physics and
Chemistry 71: 409-411.
23. ISO. Water quality - Determination of the inhibitory effect of water samples on the light emission of
Vibrio fischeri (Luminescent bacteria test). 11348-3. 2007.
24. Verordnung tiber Anforderungen an das Einleiten vor Abwasser in Gewasser (Abwasserverordnung -
Abw). Dated 21.03.1997.
25. DANETV. LUMIStox 300 Bench Top Luminometer, ECLOX Handheld Luminometer. Luminescent
bacteria test for use in waste water. Joint test plan. DHL January 2010.
26. Nelson, S.M. and Roline, R.A. (1998) Evaluation of the sensitivity of rapid toxicity tests relative to
daphnid acute lethality tests. Bull. Contam. Toxicol. 60:292-299.
27. Backhaus, T., Froehner, K., Altenburger, R., and Grimme, L.H. (1997) Toxicity testing with Vibrio
fischeri: comparison between the long term (24H) and the short term (30 min) bioassay. Chemosphere 35,
12, 2925-2938.
28. Gavaskar, A. and Gumming, L. Cost Evaluation Strategies for Technologies Tested under the
Environmental Technology Verification Program. 2001. Battelle.
29. European Commission. Commission Directive on classification, packaging and labelling of dangerous
substances. 2001/59/EC. 2001.
30. DANETV. LUMIStox 300 Bench Top Luminometer, ECLOX Handheld Luminometer. Luminescent
bacteria test for use in waste water. Joint test report DHL 2010. Published at www.etv-denmark.com.
31. DHL Internal audit of HACH-LANGE verification. Performed 22 January 2010.
32. Battelle. Quality Assurance Routing Sheet ETV Program. Joint Verification of the HACH-LANGE
GmbH LUMIStox 300 Bench Top Luminometer and ECLOX Handheld Luminometer. Audit type: TSA.
Performed 12 February 2010.
46
-------
33. DANETV. LUMIStox 300 Bench Top Luminometer, ECLOX Handheld Luminometer. Luminescent
bacteria test for use in waste water. Joint verification protocol. DHL January 2010.
34. DANETV. LUMIStox 300 Bench Top Luminometer, ECLOX Handheld Luminometer. Luminescent
bacteria test for use in waste water. Joint Test Plan. DHL January 2010.
35. Battelle. U.S. ETV Program, Advanced Monitoring Systems Center, Test/QA Plan. Verification of Rapid
Toxicity Technologies. June 11, 2003.
47
-------
APPENDIX 3
Application and performance parameter definitions
48
-------
This appendix defines the applications and the relevant performance parameters used
to verify the performance of an environmental technology following the DANETV
Program. The appendix was prepared as part of the verification protocol 7337.
1
Applications
The intended application of the product for verification is defined in terms of the
matrix, the targets and the effects of the product.
The LUMIStox and ECLOX are luminometers which measure light from the light
producing bacteria Vibrio fischeri, as indicator of acute toxicity.
1.1 Matrix/matrices
The luminometers are sold for testing of wastewater; river and lake water; leachates
from soil, waste, rubble, etc.; or directly in fluent chemicals. The matrix in which the
application is being verified is wastewater effluent from both domestic and industrial
sources.
1.2 Effect
The luminometers can measure any acute toxicity that causes an effect on the light
emission from Vibrio fischeri. In the ISO 11348-3 7237 standard, which the LUMIStox
is being tested according to, three compounds are listed as reference substances to be
included in validity testing. These are 3,5-dichlorophenol, zinc (II) as zinc sulphate
heptahydrate and chromium (VI) as potassium dichromate.
The verification will include these reference substances as well as selected metal ions,
organic pesticides, organic toxic compounds, industrial chemicals and surfactants.
1.2.1 Compounds to be tested
The vendor has suggested a list of compound to be included in the verification; these
are listed in Appendix Table 1.
Appendix Table 1 List of compounds suggested by vendor.
Group
Heavy metals
Organic pesticide
Organic pollutants
Industrial pollutant
Surfactants
Compound
Hg-complexes as HgCI2
Pb2+ as Pb(N03)2
Zn2+ as ZnSO4+7H2O
Cr2O72" as K2Cr2O7
2,4,5 Trichloroanilin
Formaldehyde
p-Crecol
Hydroquinone (benzene-1,4-diol)
Cyanide (CN-) as KCN
SDS (sodium lauryl sulphate)
CTAB (cetyl trimethyl ammonium bromide)
The vendor has performed tests on all suggested compounds except HgCb and 2,4,5-
trichloroanilin.
Each of the target groups and vendor suggested compounds was evaluated as follows:
49
-------
Hg is banned in the EU; it is therefore not likely to be found in European domestic
wastewater today. Hg is difficult to work with in the laboratory. For these reasons Hg
is excluded.
Copper is included since it is a good representative for heavy metals in both domestic
and industrial wastewater, and since it is found in wastewater as many different ions.
The ISO 11348-3 uses 3,5-dichlorophenol, Zn2+ (as ZnSO4+7H2O) and Cr6+ (as
K2Cr2O7, in water resulting in Cr2(V ) as reference substances for testing the quality of
delivered bacteria batches. Cr2(V will be included giving the possibility to do some
reference to the standard and the precision test which is described in Appendix Table 7.
Zn2+ will be included since good literature values exist.
Having two positive metals ions (Cu2+ and Zn2+), seems sufficient and Pb2+ has
therefore been excluded from the test program.
2,4,5-trichloranilin is not a regularly used pesticide. Instead a pesticide produced by the
Danish company Cheminova and included in their standard effluent wastewater
analyses is included. The specific pesticide, flutriafol, has been chosen in cooperation
with Cheminova.
Hydroquinone is not seen as a compound with special relevance for effluent
wastewater and is therefore excluded.
Formaldehyde and p-cresol are easily degradable and relatively volatile. It is therefore
unlikely that they will remain in the wastewater effluent after treatment in the plant.
Instead, triclosan, which is widely used in household products and found in domestic
wastewater, is included. Triclosan is toxic to bacteria.
U.S. EPA ETV has performed verification of similar equipment, but to be used on a
chlorinated drinking water matrix. The selection of compounds for those tests was
made with a different focus than in this verification. However, the U.S. EPA ETV
verification included cyanide, which also is included in the list of compounds
suggested by vendor. The vendor has found cyanide to be difficult to work with at pH
7. Cyanide will be included as target compound, but special actions will be taken to
ensure and monitor loss of cyanide from test solutions.
In addition to the listed surfactants, nonylphenol ethoxylate will be included in the test
since it is a well know surfactant that is very toxic to aquatic organisms and is
unwanted in the water environment. By including nonylphenol ethoxylate the three
surfactants will represent anionic, cationic and nonionic detergents.
The final list of compounds to be included in the verification is listed in Appendix
Table 2.
50
-------
Appendix Table 2 List of compounds to be included in test with notification on whether compound is
typical for domestic or industrial wastewater.
Group
Heavy metals
Organic
pesticides
Organic
pollutants
Industrial
pollutant
Surfactants
Compounds suggested by
vendor
Hg-complexes as HgCI2
Pb2+ as Pb(N03)2
Zn2+ as ZnSO4+7H2O
Cr2O72" as K2Cr2O7
2,4,5 Trichloroanilin
Formaldehyde
p-Crecol
Hydroquinone (benzene-1 ,4-diol)
Cyanide (CN-) as KCN
SDS (sodium lauryl sulphate)
CTAB (cetyl trimethyl ammonium
bromide)
Chosen compounds
Cu^+ as Cu(NO3)2
CrO72" as K2CrO7
Zn2+ as ZnSO4+7H2O
Flutriafol
Triclosan
Cyanide (CM") as KCN
SDS (sodium lauryl
sulphate)
CTAB (cetyl trimethyl
ammonium bromide)
Nonylphenol ethoxylate
Domestic
X
X
X
X
X
X
X
X
Industrial
X
X
X
X
X
X
X
X
X
X
Appendix Table 3 is a list of ECso-values for the selected compound found in the
literature.
Appendix Table 3 ECso-values from literature for the selected compounds.
Group
Heavy
metals
Organic
pesticides
Organic
pollutants
Inorganic
pollutant
Surfactants
CAS no.
7758-98-7
7778-50-9
7733-02-0
7667-21-0
3380-34-5
57-12-5
151-21-3
57-09-0
104-35-8
Compound
Cu2+ (copper sulfate)
Cr20/-
(potassium dichromate)
Zn2+
(zinc sulphate heptahydrate)
Flutriafol
Triclosan
Cyanide (CN~)
SDS
CTAB
Nonylphenol ethoxylate
EC50
(Vibrio fischeri)
mg/L
7.1
(0.35-19.5, n=3)
18.7 mg/L ±11%
2.2 mg/l ± 23%
no data found
0.28
4
2.09
0.972
no data found
According to
ISO 11 348-3
to be
determined
Yes
Yes
Yes
No
unknown
Yes
Reference
/26/
/23/
/23/
/21/
111
1221
1211
1.3 Target(s)
The targets for the application are generally reported in terms of limit of detection
(LoD), precision (repeatability and reproducibility), trueness, range of application and
robustness. For toxicity testing the limit of detection is not possible to determine.
Instead it is chosen to determine the criterion of detection (CD) based on the standard
30 minutes incubation time.
51
-------
deviation of blanks. The trueness of the inhibition is difficult to measure, and therefore
the verification of trueness will be replaced a verification of agreement with accepted
values, which will be evaluated by comparing the measured value to available robust
literature values obtained by use of the ISO 11348-3 method, for same compound. The
range of the application cannot be determined directly by identification of linear range
as for regular measurements. For this verification range is based on the inhibitions
needed to determine ECso-values, see description in Section 4 Performance parameter
definitions.
The values of the targets claimed by the vendor are given in Appendix Table 4 for the
products.
The vendor has incorporated equipment in the LUMIStox for color correction of
inhibition. With the use of the color correction on colored samples a robustness of 95-
113% was shown. Without color correction, the robustness was 109-148% 7457.
The robustness is the relative results (relative to standard) due to defined variations in
e.g. concentration level, temperature, pH, color, turbidity, cuvette types, matrix (pure
water versus wastewater). The ISO 11348-3 standard includes the possibility of testing
(marine) saltwater samples; however, saltwater samples are not included in robustness
testing of the products.
Appendix Table 4 Vendor claim of performance /5/.
LUMIStox
ECLOX
Criterion on
detection3
% inhibition
(10)
(10)
Precision (RSD)
%
Precision of
instrument
0.7
2
Precision of
test4
<20
<20
Range of
application (linear
screening range)
% inhibition
10-90
10-90
Agreement
with accepted
values
%
Not specified
Not specified
Robustness
%
Not specified
Not specified
The vendor has recently tested selected compounds. The results can be found in Table
5.1, in Section 5.1 Summary of existing data.
In the ECLOX manual the vendor states the following:
Due to nature of the simplified procedure and that the test is carried out at
ambient temperatures the results may differ if compared directly with results
[derived] for the same sample using the ISO 11348 method.
1.4 Exclusions
The verification is to be performed on one effluent domestic wastewater and one
industrial wastewater, other media are excluded. However, individual test substances
are tested in 2% NaCl MilliQ-water.
Given as part of linear range.
1 Is not clearly stated from vendor as repeatability or reproducibility.
52
-------
According to the vendor, samples containing chlorine as a result of drinking water
chlorination will interfere with the test results by affecting the viability of the bacterial
agents. Chlorine containing samples are excluded from the test.
2 General performance requirements
No formal performance requirements for the application have been identified in the
European Union or the U.S. and Canada.
The conventional performance parameters of analytical and monitoring methods and
equipment are limit of detection (LoD), precision (repeatability and reproducibility),
trueness, specificity, linearity and matrix sensitivity. The uncertainty of measurements
may be used to summarize the performance. Parameters may be added to characterize
variations of equipment, e.g. on-line or on-site monitoring instruments.
2.1 Regulatory requirements
No regulatory requirements exist for measurement of luminescent toxicity. The new
Water Framework Directive 2009/90/EC of 31 July 2009 contains a minimum
performance criteria of 25% RSD, applicable for all methods of analysis.
In Germany, wastewater regulations include results from luminescent bacteria tests
(LID, lowest ineffective dilution) as quality criteria for several industries including the
chemical industry, the rubber industry, cooling towers and waste treatment plants 7247.
For the chemical industry a LID = 32 times is accepted, meaning that the wastewater
has to be diluted a maximum of 32 times to obtain a toxicity below 20% inhibition
towards the luminescent bacteria.
For a few of the compounds, environmental quality standards for surface waters are
given by the EU 7147. These are listed in Appendix Table 5.
Appendix Table 5 Environmental quality standards stated by EU /14/and Denmark. For Denmark values
in normal writing are effective /15/, while values in italic are planned to come in force
within 2010/16/.
Group
Heavy
metals
Organic
pesticides
Compound
Cr(VI)
Cu
Zn
Flutriafol
EU
Inland surface
water
ljg/L
Other surface
water
M9/L
Denmark
Fresh water
ljg/L
4.9 (dissolved)
1.0 (dissolved)
max. 12
7.8
(dissolved)
Soft water: (H<24
mg CaCO/LJ
3. 1 (dissolved)
31
Marine water
ljg/L
3.4 (dissolved)
1.0 (dissolved)
max. 2.9
7.8 (dissolved)
3.1
53
-------
2.2 Application based needs
A validity check is required according to ISO 11348-3. The validity check involves
analysis of three reference standards which should cause 20 to 80% inhibition after 30
minutes of contact time. The results from the validity check are shown in Appendix
Table 6, as reported for the LUMIStox by vendor.
Appendix Table 6 Vendor quality data for LUMIStox according to ISO 11348-3 /6/.
Standard concentration
No. of data set
Range of inhibition
Mean inhibition
RSD
3,5 dichlorophenol
3.4 mg/L
70
22-64%
44%
27%
Zn2+
2.2 mg/l
(zinc sulphate
hepta hydrate)
60
21-49%
31%
23%
Cr2O72'
19 mg/L (potassium
dichromate)
70
48-79%
63%
11%
In ISO 11348-3, results from an interlaboratory trial with the three reference standards
are listed for information. The results are shown in Appendix Table 7.
Appendix Table 7 Interlaboratory trial, Annex C, ISO 11348-3.
No. of
laboratories
Average cone.
RSD
3,5 dichlorophenol
EC20
14
2.32 mg/L
18.6%
EC50
13
3.36 mg/L
9.6%
Zn2+
EC20
15
1 .08 mg/L
43.6%
EC50
14
2. 17 mg/L
33.6%
Cr2072-
EC2Q
15
3.60 mg/L
52.4%
EC50
14
18.71 mg/L
32.9%
3 State of the art performance
Other similar luminometers exist on the market. Some selected luminometers are listed
in Appendix Table 8. Information as to whether they have been verified is included.
Appendix Table 8 Luminometers and verification of these.
Name
Verification
Reference
Portable
BioFixLumi-10
Triathler
ToxScreen-ll
Deltatox
None known
None known
U.S. EPA ETV
U.S. EPA ETV
mi
/12/
/9/
/8/
Laboratory
Microtox
U.S. EPA ETV
111
Field installation
TOXcontrol BioMonitor
TESTNET
/10/
The three U.S. EPA ETV verifications have all been performed using drinking water
with a focus on chemical compounds toxic to humans. One compound, cyanide, is also
relevant with regards to wastewater. Performance on cyanide measurements for the
three products is listed in Appendix Table 9. The toxicity threshold is the lowest
54
-------
concentration of the tested dilutions where toxic effects were significant. For
ToxScreen-II a special set-up was used and ECso could therefore not be retrieved.
Appendix Table 9 Results from U.S.EPA ETV verification on cyanide.
Luminometer
Cyanide EC50 at 5 minutes
Cyanide EC50 at 15 minutes
Repeatability. Range of relative standard
deviation
Toxicity threshold
Microtox
8.0 mg/L
4.0 mg/L
0-4.0%
0.25 mg/L
Deltatox
7.6 mg/L
Not measured
1 .0-4.0%
0.25 mg/L
ToxScreen-II
Not measured
Not measured
0-29%
0.25 mg/L
For the TOXcontrol BioMonitor the LoD, RSD, repeatability etc. were tested and
reported for several test set-ups. The compounds used were Zn2+ and 3,5
dichlorophenol. Some of the results are summarized in Appendix Table 10.
Appendix Table 10 Results from TESTNET verification of TOXcontrol BioMonitor.
Lowest detectable change
RSD
Repeatability
Day-to-day repeatability
Memory effect
Interference (Tropaeolin-color)
Range
7.2-17% inhibition
5.7-39%
2.4-5.8% inhibition
2.5-31% inhibition
Not relevant
Not relevant
Comment
Calculated based on solution of
approximately 20%, 50% and 80%
inhibition
Calculated based on solution of
approximately 20% and 80%
inhibition
No significant effect
Increased inhibition was significant at
concentrations from 0.25 mg/L
Vendors of Vibrio fischeri test the bacteria lots and state an interval for ECso for
selected standard parameters. They also test each lot before shipment. An example of
such a test from an anonymous vendor including user laboratory reference testing is
shown in Appendix Table 11.
Appendix Table 11 ECx performance of Vibrio fischeri on standard parameters stated by vendor and
tested by vendor and user laboratory.
Standard parameter
Phenol
Zinc sulfate
Zinc2* (ion)
Specification from vendor
EC50 interval at
specification
Test time
13-26 mg/L
5 minutes
3. 0-10 mg/L
15 minutes
0.60-2.2 mg/L
15 minutes
Vendor test result
No. of LOTs
Mean
RSD
9
18 mg/L
19%
9
4.9 mg/L
27%
9
1 .0 mg/L
25%
User laboratory test result
No. of LOTS
No. of tests
Mean
RSD
9
14
18 mg/L
10%
9
15
5.5 mg/l
20%
Not tested
-
-
-
55
-------
Performance parameter definitions
Based on the above-mentioned performance requirements, a set of relevant ranges of
performance parameters for activated sludge tanks (and treated wastewater) have been
set up and are listed in Appendix Table 12.
Appendix Table 12 Relevant ranges of performance parameters in effluent wastewater.
LUMIStox
ECLOX
Criterion of
detection
% inhibition
<10
<10
Range of
application
L/L
>1/2-<1/32
>1/2-<1/32
Precision (RSD)
%
Repeat-
ability
<20
<20
Reproduce-
ability
<30
<30
Agreement
with accepted
values
%
100 ±50
100 ±50
Robustness
%
100±50
100±50
The limit of quantification is set to 10% because this is equal to the vendor claim for
linear range and because ECio-values often are used for reporting ecotoxicological
results.
The range of application for a chemical analysis is usually the range of analyte
concentration from the limit of detection to the highest concentration with linear
response. This concept is not meaningful for a toxicity test of a water sample, because
the test does not measure a concentration but an inhibitory effect as a function of the
dilution of the sample. The range of application for determining ECso therefore has to
be considered in terms of dilution. According to the HACH-LANGE manual
estimation of an ECso of a water sample requires a minimum of three measurements
where the inhibition is between 10% and 90%. In addition one of the three
measurements must be above 50%. If the standard dilution row is considered as
described in the LUMIStox 300 Operation manual and in Annex B of the ISO 11248-
3:2007 with 9 dilutions (2, 3, 4, 6, 8, 12, 16, 24, 32 times dilution in the test
suspension) then ECso should be in the range of dilutions > 2 and < 32 times dilution
assuming three measurements with inhibition between 10 and 90%. Based on test
results will be given ranges of concentrations of the compounds tested in this study,
which will give an inhibition within the range of application.
Repeatability in Appendix Table 9 and Appendix Table 10 is less than 6% in all cases,
except for the ToxScreen-II, where a repeatability of 0-29% is seen. The vendor claims
a precision for the products of < 20%, see Appendix Table 4. A repeatability of less
than 20% is chosen, since the vendor claims to fulfill this.
The day-to-day repeatability for TOXcontrol BioMonitor, as shown in Appendix Table
10, lists RSD values up to 31.2%. The vendor states, as mentioned, a test precision of <
20%, while the quality check of LUMIStox in Appendix Table 6 shows a
reproducibility of up to 27%. Here a reproducibility of 30% is chosen.
The agreement with accepted values is evaluated by looking at the ECso-values
specified by a vendor of Vibrio fischeri LOTs in Appendix Table 11. The largest
relative interval is given for zinc2+; the "mean" here is 1.4 mg/L with an acceptable
range of ± 57%. The ISO standard 11348-3 requires inhibition of 20-80% of specified
concentrations. These numbers cover both reproducibility and repeatability. The
agreement with accepted values is set to ± 50%.
56
-------
Robustness has been tested directly for the TOXcontrol BioMonitor, where the dye
chemical tropaeolin was added. The results showed a significant interference at 0.25
mg tropaeolin/L, where an increased inhibition was seen. Color correction is part of the
LUMIStox product; see section 1.3 Target(s). The robustness can be interfered by other
parameters. The general robustness is set to the level seen without color correction;
here values of 148% of true value were seen.
57
-------
APPENDIX 4
Deviation report for verification and testing
58
-------
Deviation reports
The test plan version approved must be followed. If (or rather when) deviations arc needed during tcsiinv. liw
justified in die format:
arc u
Dev.
No.
_
1 ;
t
i
|
2
-
Experiment
label
Test Plan
,
Test A
Test A
Deviation
CuS04 testing
was not com-
pleted.
Rutriafol testing-
was not com-
pleted.
Cause
Precipitation was
observed. Not toxic
at concentration with
no precipitation. Fro- i
zen samples very
different (ram fresh
samples. Inaccurate
pH adjustment gave '
different Joxicity
even when final pH
was identical.
FiutnafoJ was not
sufficiently toxic at
concentrations with
no preeiprtatton.
Impact assess- 1
ment
The Test A eval-
uation for the
heavy metal cat-
egory wtli be
based on one
fewer compound.
However, resutts
for two other
heavy mstals
MJT-
Signature
last re-
sponsible
[
V \
\\
\\
^
X^*
x3
1
^vsA
/x^
^5
\;^i
v5
XJ
Date
1
^/S
jLOtO
resporsibse
~ -(
s
;; '-
>
.%-
,<:
Vy.
vj
•".
.1"
s
'y^-
>H
'.
^
•\>
i ^
Dale i Signature '-. Oate Si^n
i Battc-iie j ETV
:: AMS QM i Can,
r • * ~' i"
! :
!
\ ''"T
V '
i
M
. ! i
-0 \ !
i N*1!) i
\ '• v vSi
IK. : sS i v
^ \
•-..: X '
J
- , ...
,
! , :•'•
i ^T-i
; 'si . i
•"•>.• i ^ ^
•i i v^ ;
•A Cs v
- i i v ^
r «v ^ i
vP
! \
\
! i
-------
05
O
3
TestD
WiH be performed
with 3 instead of 4
bacteria batches.
Oniy 3 drlterent
batches could be
delivered from Hach
| Lange.
Test on reprodu-
cibtlity is reduced
from 4 to 3 bacte-
ria batches. By
switching to three
bacteria batch
means for the
catenation of
relative standard
deviation (RSD),
the 95 paroent
confidence inter-
vai range of val-
ues for RSD is
likely to be ap-
proximately 1 5%
larger than H
would be with 4
balches. This
None T
I
j
I
was determined
by a simulation
study assuming
that the data are
normally distrib-
^ tiled.
V
-.
s>\\
N
' J
v/
\4v
\ ^^
/ \
:,W
¥
I
!
W5
SoiC
- - • — • — ~
i
i
1 >x
f\
i K>
\J
^
V '"
, V
\\!y
\4
V, \
-------
Deviation reports
The test plan version approved must be followed. If (or rather when; deviations are needed during testing, ifcc deviations are noied and
justified in the formal;
Experiment Deviation
Test Plan
Cause
Test A j Repeatability tor
i 4-NPE was not
; compieted. The
L sample was not
; sent for chemical
i analyses or Mr
croiox testing.
? Found solubility in
• literature around 1-3
j mg/i. Dissolved 3 mg
: in one liter.
Pre-screaning
showed that this
concentration was
non-toxic.
impact assess-
ment
Correc-
tive
i-action,
ii_any
Date
Signature
test re-
sponsible
The evaluation of
repeatability will
be based on few-
er compounds.
4-NPM?l deter-
gent 2 other de-
tergsflts are
present in tne
results for evalua-
tion.
Date
Signature
verificatwn
responsible
Date '
Signature
Battelle
AMS QM
Date"] Signature
ETV
I Cajiacia
-------
Otv
(W
-
o
ro
Ot
U»L plait vcrvktit approval ami be follv*ed. U (or mitei
kl (he farm*:
v Julian* are needed duiing i>*iins. tlw Jet luliuii* aic uui«-4
Plan
Tit* leal on inoVs
water
was Qtty pei-
fClTTlwd tlb Itfpll
cate. <
The t«si w
wffi«r will ba jgr
:jlt ir
non l.o*ici-
Ircipact as5*fls- ! CarrectivB
Coe luat was ac&i 1 fhe *dsie
trcr Bic 1
wst waa-dbcriad.
1 awsti sutticBni
i the toss on n •
wider.
wa-
tH
II 1«t«. ttW& FS
r>"> dfij*t^|iori a^o
1351 Will be run ni
• --p .rj v per ;!• -
anginal tesl plan
-------
. •.i-iliuri
U?
....
--.
•-1G
O
a
P,
rO
. Test
I ses^ment
j
.,.. 4 -
p*' a.x
Th.» |SO
•., !;o l
{ It « riM [Xa4nile 10
anv
Whan In*. b*C- Ir wiJ be
0! tlub bcclera W*
: &ekcd H^Cf)
: inr »ddiiionai
doe& iol ^ass itic rapvt.
aw c.'.iaria fa j hhai re-
all Sire* jtfer-
ajb-
sUrK^*s. Thai nes
dor-fl,
could
l^ provide flirc«
*'id 02099 aid rat
mfj«i «,i of Irw ISO
I ra^s* isghtiy
^ hgr.fr Miiuir-j
L-fi oil'
iei,1re.
Stftfie lh» pur-
Of r tfll O iS ttl
'• '...I' :•
Chill.(;.
w«Ui 5Q BO'i- ii
lor
Mb
StdlILL:!,. thji rp
• Ciirty
uaicties n .
ih/«« batches
. a/id the
batch
whiten did
••<•:•: mact
inctSO
Ti:-.>;.
smirf.a
and ine
[Q
^ __ ' ;ion
cri
MIL C«
Date i Shariii
t-Ji^'-ii Li-
.-
__L
••i.-
«
-------
Fin- vcnjK-,tiJL,|t pr,xi,h.--! M:p,.,ir.
m i l*e formal ;
he MJgwcxt. li
dimity te
,nul ,Usi:-
Pufflber
i
a.t
T<"i)i*&.r
TV
1 1 Th* nuir;Lit"
rfrM!h[*bt. | or
re'p.'*i-
Impaccre-
hoi, H .
Tor
The (ipijfli.ec
eqiwrens wilt
bat-
T*Q
WTI *id 4. no1 anr),-
I listed <;'i : 3 5£ Of^inally
the
. i "D updated !aa JQ- ya-
i io be iw •.•«
f C£oca
tan 1or rflpro-
Cu
3)
-------
w-Qn ; varaH»ii Vdl'lllCdlll.i';
I '.j/^-ipicr
- r.;ritf3(y.neo in
'
Appendix to a*jiajiaoi TO. a
•-• • -5pei'=:>i
? !y). *» ra-laip;* stfin-
' o*ra oeviatifin. RSD
.
i
:
j ••••gei al level i;
J c 1h* rn«ar> relalivt rgnge C
-------
Klv * 1.
' .'
.'4. •¥ *>
T"1- > «
»***-'.*r,i
S,. is E-ti'indSfCl •dfiViEfliDfl u'^ cdlasej y
I, is aftpr&g &T Ireedcrn IDJ daU&ct *
TT is Bnfe flT dalasei >•
n. ii i iLHibei o# m(?3,sur€mert[ ov aaiisrt x
nr is rlu/tiljer -3f 17183.^^111^!^ u> QnlJi^rl y
factor ift'lPrt»-£-d*d
-------
u; prtttxoi
r hjd jn thv ic-miM.
ht
rJi I] itc'ujtu.'jk-i _j[c ii^X'd^d duriu^
!. tilt
uJ i . -.
Th* wjaci«3
-------
The i*J pten
justified in th? J
muii be followed, If (or rariier
arc iK«kd ' icuin^, ihe ,-feviHtions.aie rwilcd
Dev.
We.
i
I P-i
••
: 4
I 4.3
-------
li'-ftpdiGl asses.-;
meat
- | data
i
' ~ht Ic-sL* sh^jW
t« p*rfsiTr-:-d a::-
corang lo ISO
H34S-3dT.d
12 t^nks of"DHI To*ic"o) W"
i MillQ waler j teslsd 3 sam,-
WoS
en .SL-
f'.f\90a & 'Malirj: one '.viin
v, j i.j'-:;:..
in
: =!ler!-.-.;' of 3
' simpfces SKJIIB rot
' KJ Juki ina fs-
Orm Jiasul; vnil be
Th^re
istflc! v»im a
Mijfld 1,0 die
i jih: i hr mi?
iso i i"a4&- j
redo"
|hg
i. .r-,ll_,-t:
WBl ie-
an
uwc.
Caie
'
* V ' ~
:' i V '•
:.u! ,
ETV
0#;aoii
• .,
-
A ;
1
.
, n .
-J i
. ..L....1
tS ; !
-------
i • • i.i M
I U ic* p:
•l^;:'l^il . |- :.;..- I;
appp , . ,.
-rtJ If l« nth*
i txpa«inv«[ 1 D*vaikir, j Causa
• to.
lio- •
i -4.S Ta£t«i i s"6ij^fli cnar.gbc up 53$ witn LAS.
• -•.'• Iran HKS23Q
• MS EO D£. Zl,"
' 2S" e lor anionic
... . '. i u.jffjjgt3 (as 5&Sj
SeCUOr, 2.-1 fl«w6wo("t»«ft- ~f 'iJS EPA and E°I\'
;. Ihftr experts, ly r»-
ifrit>ai =xjs»n «.rti- yiert leal ^jxxt, OWy
Ifen D.*
asm
repot si/h-
-------
I he
'.i«i in tlw: famm:
vvc a* imiril u*J j^tJ^
^ _ i__
'• Secuor Rewiw oi ;es t US EPA a^
Since one
be
. CTTV Canada (Jo i eate
Hon. il aritf
No
' tit naiuro Da!a~! S
' T^lHrnji! ' II - n'.-.f .v
i* i
•.<••
•.:
UlernJil
ETV
L
-------
^
JJfCMflCpl
CNipter
wg v
lion aLnle-
prrxluead
Mth oi *hi? r-mi
*.ictt t*Ei»d
CaTftcJi-.ni- «••-
No
i>w
two
Cato ; S«gr.fi*ure
: Dew I S*anaUife
erv
ro
------- |