Life+ FutMon - Working Group on QA/QC in Laboratories

Meeting of the Heads of the Laboratories

12-13 October 2009 in Warsaw

Limit of Detection (LOD)

and
Limit of Quantification (LOQ)
estimation and use in the chemical lab

Aldo Marchetto, Gabriele Tartari & Rosario Mosello

C.N.R. Institute of Ecosystem Study, Verbania (Italy)
e-mail: a.marchetto@ise.cnr.it http://www.iii.to.cnr.it

1
 Internal and external QC
‹ Internal quality control
¾ calibration
¾ blank control charts
¾ mean control charts
¾ LOD & LOQ
¾ result validation using ion balance andconductivity check

‹ External quality control

¾ use of certified standard
¾ analyses of certified samples
¾ participation to WRTs

2
 Why to estimate LOD and LOQ?

¾ To evaluate the suitability of the used

analytical technique and conditions to the
aims of the monitoring.

¾ To compare the quality of the determination

with other published results in order to
evaluate if there is necessity and possibility of
improvement.

3
 Aim of this presentation

¾ To show simple approaches in order to

encourage laboratories to start to estimate
their LOD and LOQ using data they still have
or they can collect at zero cost.

¾ Not to present a state-of-the-art review of

the literature about this subject.

4
 IUPAC definition

The minimum single result which, with a stated probability, can be

distinguished from a suitable blank value. The limit defines the point at
which the analysis becomes possible and this may be different from
the lower limit of the determinable analytical range.
The Limit of Detection (LOD), expressed as the concentration, cL, or the
quantity, qL, is derived from the smallest measure, xL, that can be
detected with reasonable certainty for a given analytical procedure.
The value of xL is given by the equation

xL = xbi + ksbi
where xbi is the mean of the blank measures, sbi is the standard
deviation of the blank measures, and k is a numerical factor chosen
according to the confidence level desired

5
 IUPAC definition

2002

6
 ISO 13530 1997
Water quality - Guidance on analytical quality control for
chemical and physicochemical water analysis

1997

7
 Eurachem definition

1998

8
EPA Document 815-R-05-006

2004

9
EPA Document 815-R-05-006 2004

10

10
EPA Document 815-R-05-006 2004

11

11
EPA Document 815-R-05-006
LCMRL
2004 Lowest Concentration Minimum Reporting Level

MRL Minimum Reporting Level 12

UCMR Unregulated Contaminant Monitoring Regulation

12
 Standard Methods 21° Ed. 2005: definitions

2005

13

13
 Standard Methods 21° Ed. 2005: definitions

2005

14

14
Standard Methods 21° Ed. 2005: definitions

LOD LOQ

15

15
Detection Limit in analysis
NORDTEST definition

2007

16

16
 SD of blank samples
The Limit of Detection (LOD) expressed as the concentration is derived from the
smallest measure, xL, that can be detected with reasonable certainty for a given
analytical procedure.
xL = xbi + ksbi
where xbi is the mean of the blank measures, sbi is the standard deviation of the
blank measures, and k is a numerical factor chosen according to the
confidence level desired (2 < k < 3)

To obtain LOD,
signals are transformed in
concentrations using the
calibration curve

signal 17

17
 Calibration curve
N-NH4 spectrophotometry indophenol blue
5 cm optical path

0.9

signal (A) 0.7

0.5

0.3

0.1

0.000 0.050 0.100 0.150 0.200

-0.1
concentration (mg N-NH4 /L)

18

18
 Calibration curve
N-NH4 spectrophotometry indophenol blue
5 cm optical path
0.05

0.9
0.04

0.7
0.03

signal (A)
signal (A)

0.5
0.02

0.3
0.01

0.1 0.00
0.000 0.002 0.004 0.006 0.008 0.010

0.000 0.050 0.100 0.150 0.200

-0.1 -0.01

concentration
concentration (mg N-NH4 /L)(mg N-NH4 /L)

19

19
 Calibration curve
Chloride by Ion Chromatography determination
Dionex AG19-AS19-AAES 100µL sample loop
2.0

1.5
signal (peak area)

1.0

0.5

0.0
0 0.5 1 1.5 2

-0.5

concentration (mg Cl /L)

20

20
 Calibration curve
Chloride by Ion Chromatography determination
Dionex AG19-AS19-AAES 100µL sample loop
2.0 0.20

1.5 0.15

signal (peak area)

signal (peak area)

1.0 0.10

0.5 0.05

0.0 0.00
0 0.5 0.00 1 0.05 1.5 0.10 2 0.15 0.20

-0.5 -0.05

concentrationconcentration
(mg Cl /L) (mg Cl /L)

21

21
 In practice…
If you have mean control charts at different concentration,
use them to evaluate LOD and LOQ
else, if calibration slope is very stable (i.e. spectrophotometry)
use SD of the blanks (if possible)
or of a control sample of low concentration
or of the lowest standard
else,
apply Hubax-Vos method on as many calibrations possible,
or use the variability of the calibration standards

22

22
 SD of blank samples
N-NH4 spectrophotometry indophenol blue
5 cm optical path
0.06
Mean
Mean = 0.025 mA
0.05 data
SD = 0.007 mA
Absorbance

0.04
xL = xbi + ksbi
0.03
xL = 0.025 + 3 x 0.007 = 0.046 mA
0.02
Calibration
0.01
mg N/L = 0.196 A + 0.0006
0.00
05

05

5
06

06

ot -07
7

9
06

08

t-0
t-0

-0

t-0
b-
n-

u-

r-

o-

r-
o
ar

se
ot

ap

ap
gi

ag

ag
ge

fe

In the daily determinations and in the calibration, blank mean is

subtracted from the values, then for estimating LOD
use 3SD (0.021 A), not mean + 3SD (0.046 A)

LOD 0.005 mg N/L = 0.196 x 0.021 + 0.0006

LOQ 0.014 mg N/L = 0.196 x 0.070 + 0.0006 23

23
 Variability of the lowest standard
Chloride by Ion Chromatograpy determination
Dionex AG19-AS19-AAES 100µL sample loop
0.09
Lowest standard
0.08
0.05 mg Cl/L
0.07
Peak area mean = 0.057
Peak area

0.06 Peak area SD = 0.007

0.05 xL = xbi + ksbi
for lowest standard
0.04 Mean
data xL = ksbi
0.03
xL = 3 x 0.007 = 0.021
fe 5

fe 6

fe 7
no 6

m 07

no 7
ap 6

ag 7

gi 8
08

m 9

09
ag 6

ge 8

gi 9
0

0
-0
0

0
0

t-0

0
0

-0
v-

v-

v-
o-

b-

o-
b-

u-

n-

u-
b-
r-

ag

ar
ot
no

Ratio Factor = peak area / ST concentration

RF = 0.057/0.05 = 1.14
LOD 0.018 mg Cl/L = 0.021/1.14
24
LOQ 0.061 mg Cl/L = 0.070/1.14

24
 Variability of calibration standards
Chloride by Ion Chromatography determination
Dionex AG19-AS19-AAES 100µL sample loop
15
Standard Peak RSD
mg/L area
0.05 0.058 13.8
RSD peak area

10 RSD = 5.73 (mg/L)-0.2

0.10 0.108 8.6
0.20 0.218 5.8
5
0.50 0.558 5.5
2.00 2.277 4.6
5.00 5.701 4.3
0
0 2 4 6 8 10 12 10.00 11.439 4.2
-1
Cl (mg L )
n >300 and 2-3 years of calibrations

LOD = RSD 33% = (5.73/33) (1/0.2) LOD 0.0002 mg Cl/L

LOQ = RSD 10% = (5.73/10) (1/0.2) LOQ 0.062 mg Cl/L
25

25
 Using control chart
Single control chart in use
-1
0.60
from Nov 2006 to Aug 2007 Statistics mg L
UCL
0.58 Average 0.508
0.56 UWL
Standard Deviation 0.013
0.54
Standard Deviation recommended 0.025
0.52
-1

RSD % 2.7
mg L

0.50
0.48 RSD % recommended 5.0
0.46 Minimum 0.47
LWL
0.44 Maximum 0.54
0.42 LCL
N. data 115
0.40

101
106
111
1
6
11
16
21
26
31
36
41
46
51
56
61
66
71
76
81
86
91
96
25

All control charts used in the lab for

R.S.D. Control Charts

20
IC determinations (in 20 years)
Chloride
15
Ion Chromatography
10

5 RSD = 3.37 (mg/L)-0.46

0
0 1 2 3 4 5 6 7 8 9 10 11 26
-1
mg L

26
 LOD from control charts
Chloride by Ion Chromatography determination (100µL sample loop)
25

R.S.D. Control Charts

20

15 RSD = 3.37 (mg/L)-0.46

10

0
0 1 2 3 4 5 6 7 8 9 10 11
-1
mg L

LOD = RSD 33% = (3.37/33) (1/0.46) LOD 0.007 mg Cl/L

LOQ = RSD 10% = (3.37/10) (1/0.46) LOQ 0.094 mg Cl/L
27

27
Hubaux-Vos

28

28
LOD from daily calibration (Hubaux-Vos)

Example from Dionex

Chromeleon Software
Chloride by IC
100µL sample loop
Range 0.05 – 10 mg/L
7 standards in duplicate
daily calibration

29

29
LOD from daily calibration (Hubaux-Vos)

Example from Dionex

Chromeleon Software
Chloride by IC
100µL sample loop
Range 0.05 – 10 mg/L
7 standards in duplicate
daily calibration

30
LOD 0.027 mg Cl/L

30
 LOD from average calibrations (Hubaux-Vos)
Chloride by Ion Chromatography determination
Dionex AG19-AS19-AAES 100µL sample loop
0.07
Range 0.05 – 10 mg/L
7 standards
0.05
More than 300 calibrations
0.03 in 2-3 years
Data
Signal

yc Upper prediction limit Standard Peak RSD

0.01
Upper confidence limit mg/L area
-0.01 Regression line
0.05 0.058 13.8
Lower confidence limit
-0.03 Lower prediction limit 0.10 0.108 8.6
L.O.D.
0.20 0.218 5.8
-0.05
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.50 0.558 5.5
Concentration
2.0 2.277 4.6
5.0 5.701 4.3
LOD 0.038 mg Cl/L 10.0 11.439 4.2

31

31
 Values obtained:
Chloride by Ion Chromatography determination
Dionex AG19-AS19-AAES 100µL sample loop

Results obtained (mg Cl / L) LOD LOQ

Lowest standard variability 0.018 0.061

Standards variability RSD 33/10% 0.0002 0.062

Control charts RSD 33/10% 0.007 0.094

Hubaux-Vos daily calibration 0.027 0.089

Hubaux-Vos average calibrations 0.038 0.073 *

* Gibbons’s AML 2001

32

32
 In practice…
If you have mean control charts at different concentration,
use them to evaluate LOD and LOQ
else, if calibration slope is very stable (i.e. spectrophotometry)
use SD of the blanks (if possible)
or of a control sample of low concentration
or of the lowest standard used for calibration
else,
apply Hubax-Vos method,
or use the variability of the calibration standards
Don’t calculate the LOD and LOQ for determinations of:
pH, Conductivity
Alkalinity: measured through 2 end points or Gran’s method
(with Gran’s the negative values of alkalinity indicate acidity)

33

33
 Conclusions
‹ Quality Control is an important tool to criticize analytical
activity, to improve it and to assure that it is suitable for
the purpose of monitoring.
‹ The estimation of LOD and LOQ is an important part of
Quality Control.
‹ LOD and LOQ can be simply evaluated using results of
control charts, blank samples or calibration curves.
‹ If you do not store these data yet, please store them
anywhere and in a few years you will have years of data...

34

34
 LOD & LOQ on the web
http://goldbook.iupac.org/

http://www.eurachem.org/guides/valid.pdf

http://www.chemometry.com/Research/LOD.html

http://www.chemometry.com/Research/References/ReferencesLOD.html

http://www.epa.gov/OGWDW/methods/pdfs/methods/methods_lcmrl.pdf

http://www.nordicinnovation.net/nordtestfiler/tr569_ed_3.pdf

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2556583/

35

35
 References for LOD & LOQ
A.P.H.A., A.W.W.A., W.E.F. 2005. Standard Methods for the Examination of Water and Wastewater. (Method 1030 C). Amer.
Publ. Health Ass., Washington.
Currie. L.A. 1997. Detection: international update, and same emerging di-lemmas involving calibration, the blank, and multiple
detection decisions. Chemometrics and Intelligent Laboratory Systems 37, 151-181.
Currie. L.A. 1999. Nomenclature in evaluation of analytical methods including detection and quantifcation capabilities. (IUPAC
Recommendations 1995). Analytica Chimica Acta 391, 105-126.
Currie. L.A. 1999. Detection and quantifcation limits: origins and historical overview. Analytica Chimica Acta 391, 127-134.
De Vos B., S. Lettens, B. Muys & J. A. Deckers. 2007. Walkley–Black analysis of forest soil organic carbon: recovery, limitations
and uncertainty. Soil Use and Management 23, 221–229.
EURACHEM Guide. 1998. The Fitness for Purpose of Analytical Methods. A Laboratory Guide to Method Validation and Related
Topics
Gibbons R.D. and D.E. Coleman. 2001. Statistical Methods for Detection and Quantification of Environmental Contamination.
New York: Wiley, 384 pp.
Munch D. and P. Branson. 2004. Statistical Protocol for the Determination of the Single-Laboratory Lowest Concentration
Minimum Reporting Level (LCMRL) and Validation of Laboratory Performance at or Below the Minimum Reporting Level (MRL).
EPA Document 815-R-05-006.
Mocak J., A. S. Bond, M. Mitchell, G. Schollary. 1997. A Statistical Overview of Standard (IUPAC and ACS) and New Procedures
for Determining the Limits of Detection and Quantification: Application to Voltammetric and Stripping Techniques. Pure Appl.
Chem., 69, 297-328.
Thompson M., S. L. R. Ellison and R. Wood. 2002. Harmonized guidelines for single laboratory validation of methods of analysis
(IUPAC Technical Report). Pure Appl. Chem., Vol. 74, No. 5, pp. 835–855.
Thomsen V., D. Schatzlein and D. Mercuro. 2003. Limits of Detection in Spectroscopy. Spectroscopy 18(12): 112-114.
Vogelgesang J. J. Hädrich. 1998. Limits of detection, identification and determination: a statistical approach for practitioners.
Accred Qual Assur 3: 242–255.
Voigtman E. 2008. Limits of detection and decision. Part 1, 2, 3, 4 Spectrochimica Acta Part B 63: 115–128, 128-141, 142-153,
154-165.
Janiga I., J. Mocak, I. Garaj. 2008. Comparison of Minimum Detectable Concentration with the IUPAC Detection Limit. 36
Measurement Science Review, Vol. 8, Sec. 1, No. 5, 108-110.

36