Analytical Performance Characteristic

Validation Process
• The need to validate an analytical method, as well as the procedure to
follow, are matters of professional judgment.
• Governmental and international agencies have issued guidelines for the
appropriate validation of methods, particularly for those intended for
regulatory submission.
• These guidelines generally include the following analytical performance
characteristic
1. Accuracy 5. Linearity
2. Precision 6. Sensitivity
3. Limit of detection 7. Range
4. Limit of quantitation 8. Ruggedness or robustness
9. Selectivity
• These are standard criteria set by regulatory agencies like the FDA,
ICH (International Council for Harmonisation)
 Detection Limits
The limit of detection (LOD)
• The lowest concentration level that can be determined to be statistically
different from an analyte blank.
• The concentration that gives a signal equal to three times the standard
deviation of the background above the mean background.
Determination of LOD
There are numerous ways that detection limits have been evaluated:
1. Detection based on Visual evaluation.
2. Detection based on signal to noise ratio.
3. Detection based on the standard deviation of the response and the
slope
 Detection Based on Visual Evaluation
• Applicable to both non-instrumental and instrumental methods.
• Detection Limit: The minimal concentration at which the analyte
can be reliably detected is determined by analyzing samples with
known concentrations and observing visible signals.

• Method: The detection is based on

the ability to visually distinguish the
analyte's signal (e.g., color change or
peak) from the background.
• Limitations: It is subjective and less
precise than instrumental methods,
relying on the observer's ability to
detect differences.
 Detection Based on the Signal to Noise Ratio S/N
(Peak-to-peak noise level)
This method is applicable only to analytical procedures that have
measurable baseline noise. The LOD is defined as the concentration at
which the signal of the analyte is three times the peak-to-peak noise
level or three times the standard deviation of the background noise of
the blank sample.

To estimate the LOD, http://t3.gstatic.com/images?q=tbn:ANd9GcRsqjyBZOSIrJJI-PioyHn3ULWRq9k1r0Ixi89LQHdgg6I_AKCv

Measure the signal of low-

concentration samples containing
the analyte.
Compare it to the background
signal (blank sample).
• Signal-to-Noise Ratio (S/N):
A widely accepted S/N ratio of 3:1 is used to estimate the LOD. A
S/N ratio of 2:1 was previously used but is now considered less
stringent and outdated.
• Standard Deviation (σ ):
The LOD calculation involves measuring the background (blank)
samples, calculating the standard deviation (𝜎) from the obtained
values, and determining the concentration where the signal is three
times the standard deviation (3𝜎) above the background.
• Formula: S/N​ =3×σ
Example
A series of sequential baseline absorbance
measurements are made in a spectrophotometric method, for
determining the purity of aspirin in tablets using a blank
solution. The absorbance readings are 0.002, 0.000, 0.008,
0.006, and 0.003. A standard 1ppm aspirin solution gives an
absorbance reading of 0.051. What is the LOD?
Solution
Mean = 0.004 absorbance units
The standard deviation (s) = ± 0.0032
LOD = 3×0.0032 = 0.0096 absorbance reading, above the
blank signal
 Detection Based on the Response and the Slope
The LOD can be expressed by the formula:
3.3 𝑥𝜎
LOD=
𝑆
Where:
𝜎: The standard deviation of the response (usually of the blank or
background).
s: The slope of the calibration curve.
Slope Estimation: The slope (s) is obtained from the calibration curve,
which is constructed by plotting the analyte concentration against the
measured response (signal).
• Based on the Standard Deviation of the Blank: Measure the standard
deviation of repeated blank sample readings.
• Based on the Calibration Curve: Calculate the standard deviation of
the response points from the calibration curve.
Methods to Estimate 𝜎 (Standard Deviation):
I- Based on the standard deviation 𝜎 of the blank
By analyzing an appropriate number of blank samples and calculating
the standard deviation of these responses.
II- Based on the standard deviation 𝜎 of calibration curve
Calibration curve must be obtained using samples containing an analyte
in the range of the LOD.
The standard deviation of the response can be determined based on the
standard deviation of y-intercept or the residual standard deviation of a
regression line.

Y(LOD) = YB + 3 s
YB: The response of the blank (or
background signal).
𝜎 : The standard deviation of the
calibration curve.
• Example
Calculate the LOD for analytical method that used to determine
the concentration of DDT
Y = 0.3 + 13.4 X
𝜎 = 0.085 ppm
LOD =3.3 𝜎/ 𝑠
LOD = 3.3 (0.085)/13.4 = 0.019 ppm (19 ppb)

Y(LOD) = YB + 3 𝜎
Y(LOD) = 0.3 + 3 (0.085)
= 0.56
Y = 0.3 + 13.4 X
0.56 = 0.3 + 13.4 X
X = 0.019 ppm (19 ppb)
 Limit of Quantification (LOQ)

• The limit of quantitation LOQ (or sometimes also referred as

quantification limit, QL) is the lowest possible concentration of the
analyte that can be quantified by the method in a reliable way.
Determination of LOQ
There are numerous ways that detection limits have been evaluated:
1. Detection based on Visual evaluation.
2. Detection based on signal to noise ratio.
3. Detection based on the standard deviation of the response and the
slope
 Detection based on the Signal to Noise Ratio (S/N)

LOQ can be determined using signal to noise approach where the signals
of the known concentration of the analyte is compared to that of the blank
and establishing the concentration at which it could be quantified with
some confidence.
• A S/N ratio of 10:1 is generally accepted.
http://t3.gstatic.com/images?q=tbn:ANd9GcRsqjyBZOSIrJJI-PioyHn3ULWRq9k1r0Ixi89LQHdgg6I_AKCv
 Detection Based on The Standard Deviation of The
Response and The Slope
LOQ is calculated as follows:
𝟏𝟎 𝒙𝜎
LOQ =
𝑺
Where 𝜎: the standard deviation of the response.
S: the slope of the calibration curve
There are two possibilities to determine the σ and slope for the
calculation of LOQ:
a. Based on the σ of the blank: Same as in LOD. Measure several blank
samples and calculate the σ from the obtained responses.
b. Based on the calibration curve: Using samples containing an analyte
in the range of the LOQ you must obtain and study calibration curve.
 Linearity
• Definition: The linearity of an analytical procedure refers to its ability
to provide test results that are directly proportional to the concentration
(amount) of analyte in the sample, within a specified range. In other
words, as the analyte concentration increases, the measured response
increases proportionally.
• Range: The range of an analytical procedure is the interval between
the upper and lower concentrations of analyte where the method
maintains an acceptable level of precision, accuracy, and linearity.
Evaluating Linearity: To assess linearity, a minimum of five different
concentrations should be tested, as recommended by ICH
• Acceptance Criteria: The correlation coefficient (R²), which measures
how well the data fits a straight line, should typically be greater than
0.999 to indicate excellent linearity.
Linearity
 Limit of Linearity (LOL)
• The concentration at which the calibration curve departs from
linearity. The extends from the LOQ to LOL called the dynamic range.

• A well defined relationship between detector response and analyte

concentration is crucial for quantitative analysis.

• The working range of an analysis usually defines by the linear

portion of the response versus concentration curve. The
important criterion is that the response concentration
relationship is constant and reproducible.

• Once the linear range is defined, the resulting response versus

concentration plots are often referred to as standard, calibration,
working or linear curves.
 Sensitivity
• A measure of a method ability to distinguish between small
differences in analyte concentration.
• Sensitivity is the change in signal per unit change in the amount
of analyte (it's the slope of the calibration curve)
.
• Of two methods, the one that has the steeper calibration curve will
be the more sensitive.
• The sensitivity is directly derived from a response versus
concentration plot.

• The sensitivity is the slope of the line

Δ detector response/Δ concentration.

• A very sensitive method has a large slope, so that a small change

in concentration results in a large shift in the response.
• The term sensitivity is often used in place of LOD.
Sensitivity
 Recovery
• Definition:
Recovery refers to the efficiency of an extraction or sample preparation
process, specifically the amount of analyte that is successfully extracted
from the sample matrix.
• Desirable Outcome:
High recovery of analytes is a desirable outcome, as it indicates effective
extraction and minimal sample loss. It is an important characteristic of any
extraction procedure.
• Calculation of Recovery:
𝑭𝒐𝒖𝒏𝒅𝒆𝒅 𝒄𝒐𝒏𝒄𝒆𝒏𝒕𝒓𝒂𝒕𝒊𝒐𝒏
Recovery= x 100
𝑨𝒅𝒅𝒆𝒅 𝒄𝒐𝒏𝒄𝒆𝒏𝒕𝒓𝒂𝒕𝒊𝒐𝒏
• Purpose:
Recovery can help identify if sample losses during extraction are due to
matrix effects or due to inefficient extraction. Poor recovery may indicate
issues with the extraction procedure or the sample matrix.
 Specificity and Selectivity

• Specificity refers to the ability of a method to accurately

determine the analyte level in the presence of all other components or
materials present in the analytical system (i.e., the sample matrix,
solvents used, impurities, degradation products, etc).
• Selectivity refers to the ability of a method to separate or resolve a
pair of compounds from one another (not to generate the unique
identification of the analyte from all compounds as implied
by specificity).
• Ion selective electrodes are designed to respond to one ion (e.g.,
fluoride selective electrode responds to F-) but show response to
others (e.g., hydroxide and chloride ions in this example) and
therefore are selective but not specific.
 Specificity and Selectivity

• Selectivity is a measure of a method’s freedom from interferences.

• A method’s selectivity for an interferent relative to the analyte is
defined by a selectivity coefficient, K A,I

• Selectivity coefficient (K A,I) = kI / kA

• K A,I may be positive or negative depending on the sign of kI and

kA.

• The selectivity coefficient is greater than +1 or less than –1 when

the method is more selective for the interferent than for the analyte.
 Repeatability and Reproducibility
• Repeatability: Refers to the precision of a method when replicate
analyses are performed within the same day by the same operator
using the same equipment. It measures the short-term precision of
the method.
• Reproducibility: Refers to the precision of a method over longer
periods (e.g., over a week) and under different conditions, such as
between different operators, laboratories, or instruments. It measures
the long-term precision.
• Expression of Precision: Both repeatability and reproducibility are
commonly expressed as the relative standard deviation (RSD), which
represents the variability in the measurements relative to the mean.
The RSD is calculated as:
σ
RSD = 𝑿 𝟏𝟎𝟎
µ
 Robustness
• Definition:
The robustness of a method refers to its ability to produce consistent and reliable
results when small, deliberate changes are made to the operating parameters. It
measures the method's resilience to variations in conditions.
• Determination:
Robustness is typically assessed during the method development stage and is an
important part of the intralaboratory development and validation process.
• Example of Changeable Parameters:
Organic level in the mobile phase, pH of the mobile phase, Concentration of
mobile phase modifiers, Temperature, Column type or conditions.
• Purpose:
The goal is to ensure that minor adjustments do not significantly affect the
method's performance or the quality of the results, making the method more
reliable and adaptable in real-world conditions
 Ruggedness
• Definition: The concept of ruggedness extends robustness by including
the reproducibility of a method across different analysts, laboratories, and
instruments. It assesses how well the method performs under varying real-
world conditions.
• Ruggedness Testing: It isdetermined through interlaboratory studies,
where multiple laboratories test the same sample or set of samples. This is
often done in collaborative studies involving eight or more.
• laboratories.Challenges: Ruggedness testing considers a broader range of
potential variations, including:Variations in system parameters.Differences
in sample handling, sample preparation, and mobile phase preparation.
Variability in analyst interpretation and judgment.
• Purpose: The goal is to assess the method’s reliability and consistency
when subjected to broader variations in real-world laboratory conditions.