Aldo Marchetto, Gabriele Tartari & Rosario Mosello

Limit of Detection (LOD)and

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

Life+ FutMon - Working Group on QA/QC in LaboratoriesMeeting of the Heads of the Laboratories

12-13 October 2009 in Warsaw

C.N.R. Institute of Ecosystem Study, Verbania (Italy)

e-mail: a.marchetto@ise.cnr.it http://www.iii.to.cnr.it

Internal and external QC

calibration blank control charts mean control charts LOD & LOQ result validation using ion balance andconductivity check

Internal quality cntrol

External quality control use of certified standard analyses of certified samples participation to WRTs

Why to estimate LOD and LOQ?

To evaluate the suitability of the used analytical technique and conditions to the aims of the monitoring.

To compare the quality of the determination with other published results in order to evaluate if there is necessity and possibility of improvement.

Aim of this presentation

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.

Not to present a state-of-the-art review of the literature about this subject.

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

IUPAC definition1997

IUPAC definition

2002

ISO 13530 1997Water quality - Guidance on analytical quality control for

chemical and physicochemical water analysis

1997

Eurachem definition

1998

EPA

2004

Document 815-R-05-006

EPA Document 815-R-05-006 2004

EPA Document 815-R-05-006 2004

EPA Document 815-R-05-006

2004 LCMRLLowest Concentration Minimum Reporting Level

MRL Minimum Reporting Level UCMR Unregulated Contaminant Monitoring Regulation

Standard Methods 21° Ed. 2005: definitions

2005

Standard Methods 21° Ed. 2005: definitions

LOD LOQ

Detection Limit in analysisNORDTEST definition

2007

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 inconcentrations using the

calibration curve

SD of blank samples

signal

Calibration curve N-NH4 spectrophotometry indophenol blue

5 cm optical path

-0.1

0.1

0.3

0.5

0.7

0.9

0.000 0.050 0.100 0.150 0.200

concentration (mg N-NH4 /L)

sign

al (A

)

-0.1

0.1

0.3

0.5

0.7

0.9

0.000 0.050 0.100 0.150 0.200

concentration (mg N-NH4 /L)

sign

al (A

)

Calibration curve N-NH4 spectrophotometry indophenol blue

5 cm optical path

-0.01

0.00

0.01

0.02

0.03

0.04

0.05

0.000 0.002 0.004 0.006 0.008 0.010

concentration (mg N-NH4 /L)

sign

al (A

)

Calibration curve Chloride by Ion Chromatography determinationDionex AG19-AS19-AAES 100µL sample loop

-0.5

0.0

0.5

1.0

1.5

2.0

0 0.5 1 1.5 2

concentration (mg Cl /L)

sign

al (p

eak

area

)

Calibration curve Chloride by Ion Chromatography determinationDionex AG19-AS19-AAES 100µL sample loop

-0.5

0.0

0.5

1.0

1.5

2.0

0 0.5 1 1.5 2

concentration (mg Cl /L)

sign

al (p

eak

area

)

-0.05

0.00

0.05

0.10

0.15

0.20

0.00 0.05 0.10 0.15 0.20

concentration (mg Cl /L)

sign

al (p

eak

area

)

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

SD of blank samplesN-NH4 spectrophotometry indophenol blue

5 cm optical path

Calibrationmg N/L = 0.196 A + 0.0006

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

0.00

0.01

0.02

0.03

0.04

0.05

0.06

gen-05

giu-05ott-

05feb

-06ap

r-06

ago-06

mar-07

ago-07

ott-07

apr-0

8se

t-09

Abs

orba

nce

Mean

dataMean = 0.025 mA

SD = 0.007 mAxL = xbi + ksbi

xL = 0.025 + 3 x 0.007 = 0.046 mA

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)

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

Variability of the lowest standardChloride by Ion Chromatograpy determinationDionex AG19-AS19-AAES 100µL sample loop

Ratio Factor = peak area / ST concentrationRF = 0.057/0.05 = 1.14

LOD 0.018 mg Cl/L = 0.021/1.14

Lowest standard0.05 mg Cl/L

Peak area mean = 0.057Peak area SD = 0.007

xL = xbi + ksbi

for lowest standardxL = ksbi

xL = 3 x 0.007 = 0.0210.03

0.04

0.05

0.06

0.07

0.08

0.09

nov-05feb

-06ap

r-06

ago-06

nov-06feb

-07

mag-07

ago-07

nov-07feb

-08giu-08

ott-08

gen-09

mar-09

giu-09

Peak

are

a

Mean

data

LOQ 0.061 mg Cl/L = 0.070/1.14

Variability of calibration standards Chloride by Ion Chromatography determinationDionex AG19-AS19-AAES 100µL sample loop

LOD 0.0002 mg Cl/L

0

5

10

15

0 2 4 6 8 10 12Cl (mg L-1)

RSD

pea

k ar

ea RSD = 5.73 (mg/L)-0.2

LOD = RSD 33% = (5.73/33) (1/0.2)

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

Standardmg/L

Peakarea

RSD

0.05 0.058 13.8

0.10 0.108 8.60.20 0.218 5.8

0.50 0.558 5.52.00 2.277 4.65.00 5.701 4.3

10.00 11.439 4.2

n >300 and 2-3 years of calibrations

Statistics mg L-1

Average 0.508Standard Deviation 0.013Standard Deviation recommended 0.025RSD % 2.7RSD % recommended 5.0Minimum 0.47Maximum 0.54N. data 115

0

5

10

15

20

25

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

R.S

.D. C

ontr

ol C

hart

s

Using control chart

ChlorideIon Chromatography

C.N.R.- I.S.E.Verbania - Italy

0.40

0.42

0.44

0.46

0.48

0.50

0.52

0.54

0.56

0.58

0.60

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

106

111

mg

L-1

UWL

LWL

UCL

LCL

RSD = 3.37 (mg/L)-0.46

Single control chart in use from nov 2006 to aug 2007

All control charts used in the lab for IC determinations (in 20 years)

0

5

10

15

20

25

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

R.S

.D. C

ontr

ol C

hart

sChloride by Ion Chromatography determination (100µL sample loop)

LOD = RSD 33% = (3.37/33) (1/0.46)

LOQ = RSD 10% = (3.37/10) (1/0.46)

RSD = 3.37 (mg/L)-0.46

LOD from control charts

LOD 0.007 mg Cl/L

LOQ 0.094 mg Cl/L

Hubaux-Vos

LOD from daily calibration (Hubaux-Vos)

Example from Dionex Chromeleon Software

Chloride by IC100µL sample loop

Range 0.05 – 10 mg/L7 standards in duplicate

daily calibration

LOD from daily calibration (Hubaux-Vos)

LOD 0.027 mg Cl/L

Example from Dionex Chromeleon Software

Chloride by IC100µL sample loop

Range 0.05 – 10 mg/L7 standards in duplicate

daily calibration

Range 0.05 – 10 mg/L7 standards

More than 300 calibrationsin 2-3 years

Standardmg/L

Peakarea

RSD

0.05 0.058 13.80.10 0.108 8.60.20 0.218 5.80.50 0.558 5.52.00 2.277 4.65.00 5.701 4.3

10.00 11.439 4.2

Chloride by Ion Chromatography determinationDionex AG19-AS19-AAES 100µL sample loop

LOD from average calibrations (Hubaux-Vos)

LOD 0.038 mg Cl/L

-0.05

-0.03

-0.01

0.01

0.03

0.05

0.07

0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08Concentration

Sign

al

Data

Upper prediction limit

Upper confidence limit

Regression line

Lower confidence limit

Lower prediction limit

L.O.D.

yc

Chloride by Ion Chromatography determinationDionex AG19-AS19-AAES 100µL sample loop

Values obtained:

Lowest standard variability 0.018 0.061

Control charts RSD 33/10% 0.007 0.094

Standards variability RSD 33/10% 0.0002 0.062

Hubaux-Vos daily calibration 0.027 0.089

Hubaux-Vos average calibrations 0.038 0.073 *

Results obtained (mg Cl / L) LOD LOQ

* Gibbons’s AML

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,

or use the variability of the calibration standards

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...

Thank you for your attention