Download as pdf or txt
Download as pdf or txt
You are on page 1of 15

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/51382129

Principles of analytical calibration/quantification for the separation sciences

Article in Journal of Chromatography A · July 2007


DOI: 10.1016/j.chroma.2007.03.030 · Source: PubMed

CITATIONS READS

160 5,240

5 authors, including:

Luis Cuadros-Rodríguez Maria Gracia Bagur-González


University of Granada University of Granada
158 PUBLICATIONS 4,378 CITATIONS 61 PUBLICATIONS 1,418 CITATIONS

SEE PROFILE SEE PROFILE

Antonio González-Casado
University of Granada
77 PUBLICATIONS 2,395 CITATIONS

SEE PROFILE

All content following this page was uploaded by Luis Cuadros-Rodríguez on 18 March 2018.

The user has requested enhancement of the downloaded file.


Journal of Chromatography A, 1158 (2007) 33–46

Review

Principles of analytical calibration/quantification for the


separation sciences
Luis Cuadros-Rodrı́guez ∗ , M. Gracia Bagur-González, Mercedes Sánchez-Viñas,
Antonio González-Casado, Antonio M. Gómez-Sáez
Department of Analytical Chemistry, University of Granada, E-18071 Granada, Spain
Available online 16 March 2007

Abstract
Calibration is an operation whose main objective is to know the metrological status of a measurement system. Nevertheless, in analytical sciences,
calibration has special connotations since it is the basis to do the quantification of the amount of one or more components (analytes) in a sample, or
to obtain the value of one or more analytical parameters related with that quantity. Regarding this subject, the aim of analytical calibration is to find
an empiric relationship, called measurement function, which permits subsequently to calculate the values of the amount (x-variable) of a substance
in a sample, from the measured values on it of an analytical signal (y-variable). In this paper, the metrological bases of analytical calibration and
quantification are established and, the different work schemes and calibration methodologies, which can be applied depending on the characteristic
of the sample (analyte + matrix) to analyse, are distinguished and discussed. Likewise, the different terms and related names are clarified. A special
attention has been paid to those analytical methods which use separation techniques, in relation with its effect on calibration operations and later
analytical quantification.
© 2007 Elsevier B.V. All rights reserved.

Keywords: Chemical measurement processes; Metrological and analytical calibration; Analytical quantification; Calibration schemes; Calibration methodologies

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2. Metrological fundamentals: measurement function for analytical calibration/quantification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3. Schemes for analytical calibration/quantification. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.1. Two-standard calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.2. One-standard calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
4. Methodologies for analytical calibration/quantification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
4.1. External calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.2. Matrix-matched calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.3. Standard addition calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.4. Internal calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.5. Calibration by internal normalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

1. Introduction

A chemical measurement process (CMP) is an analytical


method of defined structure that has been brought into a state of
∗ Corresponding author. Tel.: +34 958243296. statistical control, given the measurement conditions [1]. Cali-
E-mail address: lcuadros@ugr.es (L. Cuadros-Rodrı́guez). bration and validation are key operations in a CMP because they

0021-9673/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.chroma.2007.03.030
34 L. Cuadros-Rodrı́guez et al. / J. Chromatogr. A 1158 (2007) 33–46

provide traceability and comparability to the measurement, act- of an input analytical quantity, X, through the measurement of
ing the former as a metrological interface between measurement a physicochemical magnitude, Y, denoted as output analytical
standards and analytical results. Nevertheless, in analytical sci- quantity. That is why it is necessary to have a “measurement
ences, calibration can be considered from two points of view function”, Y = f(X), which permits to link the values of the out-
as metrological calibration as well as analytical calibration, not put analytical quantity with those corresponding to the input
being its requirements and aims similar necessarily. From a prac- analytical quantity. This one is a function of quantities, the value
tical view, analytical calibration implies to verify or to state of which, when calculated using known quantity values for the
the relationship between the measurement signal and the ana- input quantities in a measurement model, is a measured quantity
lyte/s quantity, whereas metrological calibration, based on the value of the output quantity in the measurement model [2].
analytical one, is just related to a “measuring system” under- In most cases, that link, always supported in a physicochem-
stood as a set of one or more measuring instruments and often ical principle, cannot be expressed in a general way by means
other devices, including any reagent and supply, assembled of a straightforward mathematical function, being necessary to
and adapted to give measured quantity values within specified use empirical models adapted to each analytical system. At this
intervals for quantities of specified kinds [2]. The metrological point, is where analytical calibration stress, due to it is the oper-
calibration implies the daily evaluation of the metrological per- ation why an empirical mathematical function is established.
formance of a measuring system in order to provide the quality Analytical calibration is always an indirect calibration, because
of the analytical results. Therefore, with a calibrated chemical is not possible to measure the quantity of a substance in a
measurement system could be possible to obtain good results system. There are different regulations where it can be found
although it provides erroneous indications, because these ones some calibration definitions, that strictly speaking are defini-
can be corrected due to, in the calibration process, the error tions of analytical calibration, which are consistent with the one
is established. That is why, a metrological calibration leads established in the most recent VIM’s version. Thus, a ISO stan-
to the characterization of a dependence relationship between dard [6] states that [analytical] calibration is a complete set of
the error committed by the measurement system and the value operations which estimates under specified conditions the cal-
of the measure, expressed in terms of parameters as devia- ibration function from observation of the response variable, y,
tion, bias, correction, correction factor or calibration factor. obtained on reference states (described by the values of vari-
This relationship is only established for the measurement val- able states, x). Also, an IUPAC recommendation [7], defines the
ues represented by the calibration standards used, but it can be [analytical] calibration as the operation that determines the func-
extended interpolating into the interval covered by them through tional relationship between measured values (signal intensities),
a correction/calibration function. Specific information about the y-variable, and analytical quantities characterizing types of
metrological approach of the calibration in a CMP can be found analytes and their amount (content, concentration), x-variable.
in references [3,4]. It can be seen that in these definitions appear the most usual
Calibration in metrology, has been recently defined in a new analytical names for the y-variable (signal intensity, response),
revision of the International Vocabulary of Metrology (VIM) and x-variable (type of analyte, analyte amount). This IUPAC
as the operation that, under specified conditions, in a first step recommendation distinguishes two different kind of calibration
establishes a relation between the quantity values with measure- functions. The first one, for species identification and qualitative
ment uncertainties provided by measurement standards and the analysis, and the second for quantitative analysis, being its main
corresponding indications [of a measuring system]1 with asso- purpose to obtain a function that allows to calculate amounts of
ciated measurement uncertainties and, in a second step, uses this an analyte as a function of an instrumental signal (quantification
information to establish a relation for obtaining a measurement function). In most cases, the calibration function has to take into
result from an indication [2] where a “measurement standard” account the response relations for all relevant constituents and
is referred to the realization of the definition of a given quantity, interferences.
with stated value and measurement uncertainty, used as a refer- These calibration functions are characteristics only for the
ence and it can be provided by a measuring system, a material standards used and depend on the actual instrumental and oper-
measure, or a certified reference material [2]. This new two- ational conditions due to the presence of “influence quantities”,
step definition generalizes the application of the previous VIM’s i.e. quantities that, in a direct measurement does not affect
definition [5] still in effect, simplifies the wording/phrasing and the quantity that is actually measured, but affects the relation
improves substantially its understanding. The main innovation between the signal intensity and the measurement results. Small
comes from the second step since as the VIM itself quoted, in changes in the influence quantity values could change the indi-
this definition the first one has been traditionally perceived as cation of the measurement system and that is why metrological
the proper calibration. As it will be highlighted in this paper, this calibration by itself does not guarantee the comparability of the
second step could include all the analytical operations closely measurements. In order to assure that the measurement system
related with calibration. remains in a calibration state, and that certain influence mag-
The aim of the chemical analysis of an object or a material, nitudes are under statistical control, some periodical controls,
is to obtain analytical information of it represented by the value named “verification” operations, are required. Verification, in
this context, is defined as the provision of objective evidence
that a given item fulfils specified requirements, taking any mea-
1 The text in brackets has been added by authors surement uncertainty into consideration [2]. The verification
L. Cuadros-Rodrı́guez et al. / J. Chromatogr. A 1158 (2007) 33–46 35

2. Metrological fundamentals: measurement function


for analytical calibration/quantification

The simplest way to write a measurement function for each


particular analytical calibration/quantification is:

Y = Φ(x, m) · X

where X and Y represent, respectively, the analyte amount (input


analytical quantity) and the analytical signal (output analyti-
cal quantity) and Φ(x, m) is the “measurement factor” which
depends on the value x of the magnitude X, and on the type of
material (matrix) m where it was. This general expression, valid
Fig. 1. Flow-chart showing up the relationships between both analytical cali-
bration, metrological calibration and analytical quantification. in all circumstances, indicates that for each analyte amount and
matrix, the analytical signal is obtained by multiplying the true
analyte amount by a specific measurement factor. Nevertheless,
is a confirmation that stated performance properties or quality the simplicity of this expression is deceitful, because in practice
requirements of a measuring system, are achieved. So, when a is impossible to characterize the Φ-parameter and to know each
measurement system has been verified, the maintenance of its one of the values it can take.
calibration state is assured. In an ideal situation, the measurement factor would just be
Thus, the main objective of calibration is to establish a characteristic of the type of analyte and independent of the ana-
mathematical function from measurement standards which sub- lyte amount and/or the matrix; in this case, the CMP could be
sequently is applied to infer analytical information from the labelled as selective. However, in analytical sciences the mea-
samples/materials analysed on the basis of an initial hypothesis suring systems show frequently low selectivity and the value
in which implicitly is admitted that, samples and standards are of the measures depend on different constituents in the mate-
metrologically equivalents. In order to obtain a quantification rial subjected to the CMP and the specific “analytical system”
free of errors, is necessary to satisfy two basic requirements: used defined as, the range of circumstances that contribute to
(i) standards and samples composition must be as similar as the reliability of analytical measurements, including measure-
possible and (ii) standards and samples must have an identical ment system, reagents, procedures, test materials, personnel,
behaviour in the measurement system. In any case, the standard environment and quality assurance trials (adapted from [8]).
must be representative of the sample; usually the first situation The “selectivity” of a measuring system is its capability, using
although desirable, is not essential being enough to fulfil the a specified measurement procedure, to provide measurement
second requirement to assure representativeness. The interre- results, for one or more measurands (quantities intended to be
lation between both analytical calibration and quantification is measured), that do not depend on each other nor on any other
arranged in sequence in Fig. 1. quantity in the system undergoing measurement [2]. Actually,
In general, the representativeness is satisfied using standards the analytical signal depends on: (i) the properties of the ana-
of the analyte in the very material to be analysed. Nevertheless, lyte species involved; (ii) the instrumental parameters of the
in those separation methods in which a multi-analyte quantifi- particular equipment; (iii) the experimental conditions (mostly
cation is going to be done, the calibration for each separate adjusted by the analyst); and (iv) the presence of accompany-
analyte can be tedious or even impossible (if there are no ing substances [9]. In order to convert the previous expression
available standards), being necessary to take into account other in other one more accessible from a practical point of view, to
options. improve the selectivity of the CMP, minimizing the lack of selec-
In this paper, the necessary conditions to assure the represen- tivity of the measuring system, some different strategies can be
tativeness and the strategies that can be adopted to minimize the applied. One of them, is to use a separation technique previ-
errors in the quantification when this one was not fulfilled, are ous the measurement step in order to: (i) remove or diminish
going to be established. Thus, the metrological fundamentals of the matrix constituents (clean-up) to do the Φ-parameter inde-
the calibration/quantification, the calibration schemes depend- pendent from the rest of the constituents of the material, and
ing on the number of standards employed and the methodologies (ii) to separate the sample in fractions (in a continuous or in
that can be used for the calibration/quantification in separation a batch way), to obtain only an analyte in each one to assure
sciences, will be widely explained, emphasizing the practical that the indication provided by the measuring system is only
requirement for making measuring and discussing common mis- attributed to the very analyte. This meaning of selectivity term
conceptions and errors that might arise. Consequently, names is coherent with IUPAC definition [10]: the extent to which the
as external calibration, internal calibration, internal normaliza- method can be used to determine particular analytes in mix-
tion calibration, matrix-matched calibration, standard addition tures or matrices without interferences from other components
calibration, and signal-ratio calibration together with terms as of similar behaviour.
external standard, surrogate or internal standard will be dis- As a consequence of this fact, the general expression of the
cussed and clarified. measurement function above described, could give rise to dif-
36 L. Cuadros-Rodrı́guez et al. / J. Chromatogr. A 1158 (2007) 33–46

ferent mathematical expressions. Danzer [11] analyses a general ical knowledge concerning the analytical system in question,
functional relationship adapted to the special features of a CMP bearing in mind that, in a lot of cases it is very difficult to obtain
in analytical chemistry, discussing the variables that affect the an analytical blank for the overall CMP (i.e. a sample that is iden-
value of the analytical signal and the consequences over some tical to those being taken for analysis but containing none/not
important characteristics of analytical methods. detectable levels of the analyte(s) sought). When the absence
One of the easiest way to describe the measurement function of a matrix effect is verified, this blank can be substituted by a
for an analyte (not taken in consideration the matrix) is: reagent or procedural blank [12] carrying out a complete anal-
ysis using the solvents and reagents only, in the absence of any
Y = Y0 + Yanal = Y0 + S(x) · X sample.
which indicates that the value of the analytical signal depends “Sensitivity”, has been defined as the quotient of the change in
on two components: one, independent of the quantity of ana- the indication and the corresponding change in the value of the
lyte, denoted as blank signal (Y0 ) and the other, the net signal quantity being measured [2]. It denotes the quantity of analytical
generated by the analyte (Yanal ). This last is expressed as a signal yielded by each unit of analyte amount in the measured
product of two terms, where S(x) is the sensitivity of the material and it can be constant or dependent of the very analyte
measurement system (not necessarily a constant because its amount. It would be desirable that sensitivity could achieve the
value can be dependent on the actual x-value), and X, the highest value possible in every CMP, in order to obtain the best
amount/concentration of analyte. This expression belongs to the performance characteristics for the analytical method implied.
analytical calibration function obtained in the validation step of From a calibration experiment appropriately designed in
the CMP, when the system is fitted to a linear response. Thus if which the analytical signal is measured from the measurement
this equation is rearranged in the way: standards, the measurement function is explicitly stated as a
  representative calibration function, which can be written in a
Y0 general way as:
Y= + S(x) · X
X
y = f (x)
the term in brackets will be the measurement factor:
Usually it is obtained fitting the different (x, y) pairs, by applying
Y0
Φ= + S(x) a proper regression algorithm. The calibration features, blank
X signal and sensitivity, are estimated by the independent term
The blank signals may be described by three different terms, and the gradient, respectively, and they are identified with the
depending upon their origin as Currie stated [1]: (i) the intercept and the slope when the measurement function is plotted
instrumental background. The null signal (which for certain as a linear calibration curve.
instruments may be set to zero, on the average) obtained in the For any standard analyte amount, the measurement factor
absence of any analyte- or interference-derived signal; (ii) the could be estimated by mean of a “response factor” (RF) from
(spectrum or chromatogram) baseline. It comprises the summa- the following equation:
tion of the instrumental background plus signals in the analyte y
(peak) region of interest due to interfering species; and (iii) the RF =
x
analyte blank. This signal arises from contamination from the
reagents, sampling procedure, or sample preparation steps which To carry out the quantification of an analyte in a sample, the
correspond to the very analyte being sought. In our opinion, the corresponding output analytical magnitude (y-variable) is mea-
blank signals are produced by: (i) the instrumental background; sured and a numerical value of the input analytical magnitude
(ii) the signal in the analyte peak region due to other accompa- (x-variable) is calculated from the inverse calibration function,
nying species which can be exogenous, e.g. from the reagents denoted as “evaluation function” [1,7] or “analytical function”,
(the so-called reagent blank), or endogenous, e.g. from interfer- according to the following expression:
ents or matrix sample (the so-called matrix blank); and (iii) the xspl = f −1 (yspl )
signal due to the presence of the very analyte in the reagents.
The combination of instrumental background and reagent blank where f−1 is the inverse calibration function and the subscript
constitutes the method blank. “spl” indicates the sample. In a similar way, each analyte amount
The blank signal origins a (non-analytical) net signal (it in the sample could be characterized by means of a “calibration
can be considered constant for each sample in each analyti- factor” (CF), defined as:
cal system) which is independent of those signals attributed to xspl 1
matrix–analyte interactions, closely related with the existence CF = =
yspl RF
of a matrix effect. Ideally, when a chromatographic method is
applied, the blank signal should be zero, since the analyte must Thus, this feature, could be used to estimate the actual analyte
pass through the measurement system separated from the other amount in a sample according to:
components of the sample and the instrumental background also
xspl = CF · yspl
may be set to zero.
The assessment of the blank (and its variability) is mandatory Table 1 shows the most frequent possibilities of calibration
for accurate low-level measurements, but require expert chem- curves as well as the RFs and CFs associated to them, being the
L. Cuadros-Rodrı́guez et al. / J. Chromatogr. A 1158 (2007) 33–46 37

Table 1
More frequent calibration curves and different features for the analytical separation sciences
Calibration curve Calibration function Signal blank Sensitivity Response factor

Linear y = bx 0 b [constant] b [constant]


y = a + bx a b [constant] a/x + b [depends on x]
Parabolic y = cx2 0 2cx [depends on x] cx [depends on x]
y = a + cx2 a 2cx [depends on x] a/x + cx [depends on x]
y = a + bx + cx2 a b + 2cx [depends on x] a/x + b + cx [depends on x]

linear and parabolic calibration functions the most frequently on the calibration curve, being the estimation of the calibration
used in analytical separation sciences. function strongly affected.
As it can be seen, in the linear functions sensitivity is constant For multi-analyte analytical methods, a very common situ-
and therefore independent of the amount of analyte. Never- ation when continuous separation techniques are applied, the
theless, the response factor is only constant when the blank calibration standards are a set of mixtures containing the differ-
signal is null. In these circumstances, sensitivity and response ent analyte standards to be quantified in the adequate amounts.
factor are equivalent and the calibration factor is the inverse The multi-standard is feasible when the number of analytes
of each one of them. Thus, the verification of these two con- allows it and analyte measurement standards are available. The
ditions (linear response and null blank signal) is crucial in measurement analyte standards used as “calibrants” (a measure-
order to avoid errors associated with the selection of certain ment standard specifically used in calibrating, also so-called
quantification strategies based in the use of a only measure- “calibrators” [2]) in analytical sciences, are reference materials,
ment standard for calibration purpose, as will be pointed out in that can be essentially classified in two types [22]:
Section 4.
(1) Substance reference materials (substance RM), prepared
3. Schemes for analytical calibration/quantification from single analytes. They include: (a) pure substances char-
acterized for chemical purity and/or trace purities, and (b)
The multi-standard calibration [13] is the usual scheme for standard solutions, prepared from pure substances (singles
the analytical calibration and the performance characteristic or mixed) in the working solvents used.
establishment of a CMP. It is based on the measurement of a (2) Matrix reference materials (matrix RM) characterized for
calibration standards set, including the blank if this is manifest, the composition of specified major, minor and trace chemi-
evenly spaced over the analytical method working range and cal constituents. They may be prepared from the matrices to
preparing replicates of each one in an independent way. Then, be analysed or by mean of synthetic mixtures. Thus, this RM
applying an adequate regression algorithm (not necessarily the contains the analytes of interest plus the principal chemical
Gauss least-square fitting method) to the data, a calibration compounds characterizing the matrix to be matched.
function is obtained. These days, new regression methods e.g.
robust regression, are been taken into consideration. A regres-
sion involves two steps: (i) selecting a mathematical model In relation to the number of calibration standards which must
(measurement function) and fitting the corresponding mathe- be used, and the number of replicates at each calibration level,
matical function from the experimental data; and (ii) validation different recommendations in recognized written standards and
of the model and checking of certain initial hypotheses. The guidelines can be found. So, IUPAC advises for method val-
consistency of the regression results always depends on the idation purposes, the use of six or more calibration standards
grade of the experimental random errors. Therefore, if these that should be run, preferably triplicate on more, in a random-
ones are important, the conclusions of the applied statistical ized way [23]. In the same way, the ISO 8466 states, for an
tests are not reliable; for calibration purpose, this fact has spe- initial assessment of the calibration performance, to employ at
cial relevance in relation to linearity or goodness-of-fit tests. It least five calibration standards, although it recommends 10, and
must be taken into account that regression is only a statistical 10 replicates of the lowest and highest standards [14]. Another
tool useful to obtain/estimate calibration functions, being other schemes with a non-balanced uniform-level design for calibra-
strategies also possible. To deal with the statistical approach tion purposes have been recommended [24].
of the calibration, out of the aims of this paper, there are These severe requirements are tempered when the analyt-
available numerous guidelines [7,14–16] and technical papers ical method is running in a routine way, being enough with
[17–21]. three or four calibration standards, duplicated if possible; this
In order to assure the accuracy of the results in samples with short-working calibration for routine analysis is termed as “con-
low analyte content, i.e. to avoid a leverage effect on the cali- tinuing calibration” [13]. If the calibration curve has been well
bration curve, it is advisable to use a homogeneous arrangement established during method validation and a suitable calibration
of the calibration standards. Thus, for example, if in a 5–100 control scheme is put into practice, it may be even possible to
analyte amount arbitrary units, the levels chosen are 5, 25 and use two alternative schemes: (i) two-standard calibration and (ii)
100, the highest standard will have a considerable leverage effect one-standard calibration.
38 L. Cuadros-Rodrı́guez et al. / J. Chromatogr. A 1158 (2007) 33–46

3.1. Two-standard calibration a low test samples number with the aim to minimize the total
analysis time. Whether the number of test samples is great, a
When the linearity of the calibration function has been batch processing (of them) can be done.
ensured in the validation step, a routine calibration can be
obtained from only two calibration standards, preferably mea- 3.2. One-standard calibration
sured in replicate. The standards must be selected in such way
that the sample analyte/s quantity be included in the covered The calibration could be performed from only one calibrant
range by them. That is why the quantification can be done inter- when two conditions are fulfilled: (i) the calibration function
polating between both standards values, e.g. by applying the must be linear in the interval of analyte amount ranged from
next equation: the standard value to zero, and (ii) the blank signal must be
x2 (ȳspl − y1 ) − x1 (ȳspl − y2 ) null in the interval previously referred to. So, it is mandatory
xspl = to check this last requirement before this calibration scheme
y2 − y 1
was applied. During the experiment, the calibrant must be repli-
where x1 and x2 are each one of the two standards analyte cated twice at least. The quantification is carried out using the
amounts, y1 and y2 the arithmetic means of the corresponding calibration factor calculated from the pair of values, standard
analytical signals, and ȳspl is the arithmetic mean of the analyte analyte amount/average analytical signal, concerning the cali-
analytical signal in the sample. Also, quantification can be done brant applying the expressions:
using the expression of the straight line through two points (both x1
calibration standards) as follows: xspl = CF · ȳspl , where CF =
ȳ1
ȳspl − ȳ1,2 ȳ2 − ȳ1
xspl = + x̄1,2 , where b = The linearity requirement can be avoided when the analyte
b x2 − x 1 amount in the selected calibration standard is very near to the
being ȳ1,2 and x̄1,2 the global means of the two measured one in the sample which is possible when the features of the
analytical signals and analyte amounts of the two standards, sample are previously known. Then the one-standard calibra-
respectively, and b is the slope. tion can be easy going applied (for example, for quality control
When it is known that the intercept is zero, the calibration of manufactured products).
factors of both standards are the same and very similar to the As for the preceding case, for pesticide residue analysis, it has
inverse of the slope. Then, the quantification could be easily car- been stipulated that one-standard calibration may provide more
ried out calculating a pooled calibration factor, CF1,2 , according accurate results than multi-standard calibration if the detector
to the expressions: response is variable with time [12]. In this case, it must be tak-
ing into account that, unless further extrapolation is supported
x̄1,2 1
xspl = CFpool · ȳspl , where CFpool = CF = = by evidence of acceptable linearity of the calibration function,
ȳ1,2 b the sample analytical signal should be within ±10% of the cali-
The two-standard calibration can be advisable, when the bration standard analytical signal if the maximum residue limit
expected range of analyte amount in the samples to be analysed is exceeded; if it is not exceeded, the sample response should be
is narrow, but it could be a very laborious calibration scheme within ±50% of the calibration response.
when the samples is unknown. For pesticide residue analysis in
particular, the use of two-standard calibration has been stated 4. Methodologies for analytical
to be acceptable providing the two calibrant x-value differs in a calibration/quantification
factor not greater than 4, and where the mean response factors,
derived from replicate determinations at each standard, does not In order to the calibration function, previously established
vary in more than a ±20%. This is indicative of an acceptable from standards, could be applied to the samples (Fig. 1), differ-
linearity of the observable analytical signals [12]. ent calibration methodologies [13] can be used, bearing in mind
A particular type of two-standard calibration is the one factors like: (i) the availability of representative standards; (ii)
labelled “bracketing calibration” [15,16]. The basis of this strat- the analytes nature; (iii) the lack of precision or drift of the mea-
egy is to decrease, as much as possible, the interval for which the surement system; and (iv) the possibility of disturbance caused
calibration function is linear, delimiting it with two calibrants in the output analytical signal by other components concomi-
which “bracket” the analyte amount values of the test samples tants the analyte, considered them individually (interferent) or
closely; both calibrants and samples, should be measured at least in a whole way (matrix effect). Obviously any of the previously
twice. Assuming this fact, bracketing calibration can be applied mentioned schemes: multi-, two- or one-standard calibrations
when there is some doubt about the linearity of the calibration can be applied in all the methodologies.
function in the range where the samples are measured and it is The calibration standards are prepared from RMs containing
specifically useful when the quantification of analyte amount in the analyte or a “surrogate”, that is, a pure substance (com-
samples is required with a high accuracy degree (in terms of pound or element) similar to analyte of interest in chemical
trueness and precision), e.g., in doping control analysis. composition, separation and measuring which is taken to be
It can be also convenient when the measurement system representative of the native analyte; this must be absent or in
shows a significant drift in its detector response, being necessary a negligible initial concentration in the sample. The calibration
L. Cuadros-Rodrı́guez et al. / J. Chromatogr. A 1158 (2007) 33–46 39

standards can be measured in a preparation apart from the sample extended to capillary electrophoresis, in order to maintain
(external methodology) or in one including the sample (internal the injection required precision during long routine injection
methodology). Usually, a surrogate is used in an internal method- sequences and/or analysis of samples in complex matrices. In
ology and, in this case, it is termed as “internal standard” (IS). fact, many imperceptible factors affect the volume introduced
The IS is chemically distinct from the analytes and therefore into the capillary during a pressure injection in CE. These fac-
will not have identical chemical properties. However, it is usu- tors, mainly related to variability in the pressure/timing of the
ally selected in such way it was closely related with the analytes injection and to changes in sample solution viscosity, can be
and can represent their analytical behaviour to the highest degree efficiently eliminated by means of the signal-ratio calibration
practicable. The most favourable way to apply IS, is to use an [25].
isotopically labelled analyte, so the chemical properties of the The signal-ratio calibration can also be used to correct a mod-
internal standard are virtually identical to the very analyte. Both, erate proportional (no-additive) matrix effect (see Section 4.2),
the very analyte and the labelled one, can be measured separately since the premise is that, a matrix effect changes the signals from
by mass spectrometry (this principle is the basis of the isotope analyte and the internal standard in the same degree, so that cal-
dilution mass spectrometry (IDMS) technique). ibration on the ratio signals against analyte amount, will remain
In calibration, the IS can be applied with two aims: (i) to invariant to changes in matrix (different composition between
estimate a calibration feature as response factor or calibration the calibration standards and the sample analysed).
function (it will be the subject of a detailed comment and it will Ideally, the standard should be added exactly to the test
be pointed out further on Section 4.4); and (ii) to compensate dissolution before the injection, in enough amount so that its ana-
uncontrolled analytical signal variations in the measuring sys- lytical signal be equal or higher than the highest analyte signal
tem. This strategy, in our opinion is not properly a methodology expected; thus, the signal-ratios will always be lower than 1, min-
for calibration, and should be called “signal-ratio calibration” imizing the imprecision of these signals and ensuring normally
although the general expression to refer to it is “internal standard distributed measurement errors [26].
method”. The distinctive features of the different methodologies,
The main characteristic of the signal-ratio calibration is based explained in next subsections, are summarized in Table 2, in
on the use of a relative signal (signal-ratio), defined as the quo- which the minimum number of preparations for a metrological
tient between the signal attributed to the analyte (analyte signal) calibration/quantification are included.
and to the internal standard:
ya 4.1. External calibration
yR =
yIS
The external calibration (EC) is the most commonly
where the subscripts “a” and “IS” are related to the analyte and employed calibration methodology and it is named so, because
the internal standard. the calibration standards do not make up of the sample test
It has been widely used to correct mechanical losses of portion but they are prepared and analysed separately from the
analytes in procedures such as manual injection in gas chro- samples. In this sense, standards and samples form part of differ-
matography. Nowadays, the use of internal standards has been ent analytical preparations which are measured in a sequential

Table 2
Types and main distinctive features of different calibration methodologies
40 L. Cuadros-Rodrı́guez et al. / J. Chromatogr. A 1158 (2007) 33–46

way. In bibliography, EC has been also termed as “solvent cal- ya,spl


xa,spl = CF · ya,spl = xa,std ·
ibration”, “standard calibration” or even “normal calibration”. ya,std
As external standard (ES), a solution of a RM substance in the
When the signal-ratio calibration is applied, it can be established
working solvent is used. It usually contains the analyte, although
a relative response factor (RRF) or a relative calibration factor
it is not essential, and the calibration could be carried out from a
(RCF) which are calculated from the ratio signals. The equations
surrogate. The use of a surrogate in external calibration is only
used are:
appropriate when there are no a RM containing the analyte. An
application, in which two surrogates are used as ES, can be found RF ya,std /xa,std ya,std /yIS,std xIS,std R
RRF = = = = · ystd ;
in a recent paper on the determination of polyphenols in olive RFIS yIS,std /xIS,std xa,std /xIS,std xa,std
oil by CE-UV [27]. 1
When a multi-standard calibration is performed, the linear RCF =
RRF
measurement function to estimate from data regression is:
Y = Y0 + S · X ya,spl
xa,spl = xIS,spl · RCF · = xIS,spl · RCF · yspl
R
which give rise to a calibration curve: yIS,spl

ya,std = aEC + bEC · xa,std It is habitual, but not essential, to add the same amount of inter-
nal standard to all the analytical preparations (calibrants and
where the subscript “std” are related to calibration standard, the samples). Then, due to xIS,std = xIS,spl , both terms can be elim-
intercept aEC is the method blank (Y0 ) and the slope bEC is the inated in the previous equation and a pseudo-relative response
sensitivity of the measurement system (S). The quantification factor (RRF ) or pseudo-relative calibration factor (RCF ) can
can be then performed from: be calculated according to:
ya,spl − aEC
xa,spl = ya,std /yIS,std yR 1
bEC RRF = = std ; RCF =
xa,std xa,std RRF
where the subscript “spl” are now related to sample. In cer-
tain cases, a no-linear calibration function, as polynomial, yspl
R
ya,spl
exponential, sigmoidal, etc., could be applied.When a signal- xa,spl = RCF · = RCF · yspl
R
= xa,std · R
ratio external calibration is performed, the calibration functions yIS,spl ystd
should be established from signal-ratios; e.g. for a linear func- EC is suitable for the analytical methods that could be considered
tion, the equation to be fitted is: as “matrix-effect free methods” (the so-called Type II methods
ya,std   in ISO Guide 32 [30]). But it have a main limitation that derives
= ystd
R
= aEC + bEC · xa,std
yIS,std from the assumption that, the differences between the matrices
of the standards (usually a working solvent) and the sample have
and the quantification equation is:
no effect on the calibration function, bearing in mind that if these

(ya,spl /yIS,spl ) − aEC differences are ignored, additive and/or proportional systematic
xa,spl = 
bEC matrix errors may be introduced. In general, the EC methodol-
ogy is recommended if the analyst has a good knowledge of the
To do this, it is mandatory that the sample test portion and all
experimental conditions and the contribution of interference to
the calibration standards were prepared with the same internal
the measurement is irrelevant or can be kept constant; neverthe-
standard amount.
less, in the last case, a proper correction of the matrix should be
A representative example of this procedure would be
applied [31], for example, by performing a “correction function”
the1613 EPA method for the determination of tetra- through
[32].
octa-chlorinated dioxins and furans by isotope dilution
high-resolution capillary column gas chromatography/high-
4.2. Matrix-matched calibration
resolution mass spectrometry [28]. In it, a five-standard external
calibration is used in order to estimate the linear signal-ratio
The matrix of calibrants and sample test portion never are
calibration function used for quantification purposes. A 13 C
exactly equal, although the sample has been previously subjected
isotopically labelled dioxine is used as internal standard.
to a clean-up procedure. Sometimes, this fact is irrelevant and
In routine analytical determinations, of the use of response
any errors in the determination are produced, but in other cases,
and/or calibration factors obtained from one-standard calibra-
a systematic error appears due to the matrix (matrix systematic
tion is also quite applied, being it advised too in some official
error) not being possible then to apply external calibration.
analytical method, for example, the Harmonised Methods of the
In separation techniques, this matrix effect is due to the influ-
International Honey Commission for determination of sugars in
ence on the measurement of the analyte amount, of one or
honey by HPLC or GC [29]. In this case, the equations to apply
more undetected sample components and derive from various
are:
physical and chemical processes. It tends to be variable and
xa,std 1
RF = ; CF = unpredictable in occurrence, being in some cases, difficult or
ya,std RF impossible to eliminate. The presence or absence of such effect,
L. Cuadros-Rodrı́guez et al. / J. Chromatogr. A 1158 (2007) 33–46 41

may be demonstrated by comparing the response produced from


the analyte in a working solvent solution with that obtained from
the same amount of analyte in the presence of the sample or sam-
ple extract. Usually, the matrix produces a change in the signal
which can be: (i) constant and independent of the analyte amount
(additive or translational effect); (ii) variable and proportional
to that amount (proportional or rotational effect); or (iii) the
combination of both type of effects [31,33].
Whether matrix problems are suspected (Type III methods in
ISO Guide 32 [30]), more reliable calibration may be obtained
using as calibration standard, a matrix RM (when available)
or a pure substance RM in conjunction with free-analyte sam-
ple matrix prepared freshly (most usual). This is the so-called
“matrix-matched calibration” (MC) and may make up for matrix
effects although it does not eliminate the underlying cause. That
is why, the intensity of effect may differ from one matrix or
sample to another, and can be also affected by the “concen-
tration of matrix”. In fact, a MC is a particular type of EC in
which the calibration standards are prepared in a simulated sam-
ple that initially it does not contain analyte. Nevertheless, the
MC will not overcome chromatographic interferences caused by
overlapping/unresolved peaks from co-extracted compounds. In
addition, when matrix effect is sample dependent it is necessary
to use standard addition calibration (see next section).
Metrologically, the matrix effect origins a modification in Fig. 2. A particular example of the relationship between different types of cali-
the measurement function, due to the appearance of new terms. bration curve plots: external calibration (EC), matrix-matched calibration (MC),
Supposing a linear behaviour, it can be represented as: standard addition calibration (AC) and empirical matrix-matched calibration or
matrix-corrected calibration (MCC ). y0 , method blank signal; ym , matrix blank
Y = Y0 + Ym + Yanal = Y0 + Ym + p · S · X signal; ya,spl , signal from actual analyte in sample; xa,spl , actual analyte amount
in sample; aEC , intercept from EC-curve; aMC , intercept from MC-curve; aAC ,
where Ym indicates the matrix sample contribution to the blank, intercept from AC-curve; aYC , intercept from Youden calibration curve (or YC-
i.e. the matrix blank (this term characterizes the additive matrix curve); bEC , slope from EC-curve; bMC , slope from MC-curve; bAC , slope from
effect and can have positive or negative values) and p character- AC-curve.
izes the proportional one (It can reach values upper and down 1)
being it a feature defined in a similar way to the so-called “selec- If is necessary, the MC-function also could be performed from
tivity index” in IUPAC nomenclature [23]. It can be expressed signal-ratio calibration by using a suitable internal standard,
as: being the equations to apply:
Sm ya,std  
p= ; and so : sm = p · S = ystd
R
= aMC + bMC · xa,std
S yIS,std

where Sm is the analyte measurement system sensitivity when 


(ya,spl /yIS,spl ) − aMC
the matrix sample is present. xa,spl = 
bMC
The presence of a significant matrix effect brings about a MC-
function different to that corresponding to EC-function which This methodology is particularly recommended in pro-
for a linear function, can be stated as: cedures for pesticide or drug residue analysis and other
contaminants in food and biological matrices [34,35]. Contrary
ya,std = aMC + bMC · xa,std to the common belief, the reliability of quantification using
where the intercept and the slope can be expressed as: modern analytical techniques as GC–MS or even LC–MS (or
aMC = aEC + ym = y0 + ym , and bMC = pS = pbEC . It can be tandem MS/MS) might not be adequate. Results can be adversely
deduced that, when ym = 0 and p = 1, there is not a matrix effect affected by the lack of selectivity caused by matrix sample effect
and both calibration functions are equivalents. A particular case on ion suppression.
in which ym > 0 and p > 1 is shown in Fig. 2, as an example. Owing to the variety of matrices and samples in pesticide
In the laboratory, the sample must be prepared in such way residue analysis, the MC-functions are prepared only in a series
the sample amount in the test portion was similar to the matrix of representative matrices, this is, a sample material or an extract
amount used to prepare the calibration standards. of a single food or feed used to represent a commodity group as
an indicator of matrix effect in the analysis of broadly similar
ya,spl − aMC commodities. Matrix representativeness is usually determined
xa,spl =
bMC among similar biological material (or tissues) according to its
42 L. Cuadros-Rodrı́guez et al. / J. Chromatogr. A 1158 (2007) 33–46

content in constituents like, water, acids, sugars, lipids, sec- considered a “true” calibration, since no measurement standards
ondary plant metabolites, etc. The utility of these representative are used, but just increasing sample amounts. The YC-curve is
matrices, has been demonstrated preparing MC calibrations in obtained from a plot of the measured signal against the sample
cucumber extracts for the determination of 20–30 pesticides by (not analyte) amount (xspl ):
GC–MS [36] or LC–MS [37] in several vegetable samples.
ya,spl = aYC + bYC · xspl
4.3. Standard addition calibration and the corresponding intercept (aYC ) is the total sample blank
(so-called total Youden blank) [33,39]. Consequently the YC-
The establishment of a matrix-matched calibration needs a intercept, in agreement with the MC-intercept (see Fig. 2a and
matrix free from analyte which it is only possible when the b), is a measure of the total sample blank:
analytes are exogenous sample components (e.g. additives or
contaminants or additives). Nevertheless, if the analytes are aYC = aMC = aEC + ym = y0 + ym
endogenous (e.g. aminoacids in biological fluids or pigments and the quantification equation is:
in vegetables), and a matrix effect has been verified, it is neces-
aAC − aYC
sary to apply the method of standard addition, properly named xa,spl =
“standard addition calibration” (AC). There are different ways in Sm
which the AC can be applied which have been recently reviewed When the total sample blank is null, the former equation can be
[38]. simplified to:
It is a particular internal calibration methodology, usu- aAC
ally based on the analysis of several test portions made, xa,spl =
bAC
adding increased amounts of analyte standard to several sam-
ple aliquots. From these data, an AC-function can be obtained, which gives the extrapolated value of the x-intercept. Only in
which will have the same slope as the MC-function but higher this case, the analyte amount can be calculated as the quotient
intercept, assuming that it can be considered as a MC in which between the intercept and the slope of the AC-curve. This fact
the matrix contains analyte. Fig. 2a shows a linear AC-curve is the basis of the graphic quantification method that generally
together with the corresponding linear MC- and EC-curves. appears in text books. Nevertheless, if the total sample blank
In this case, the measurement linear function is expressed by nullity is not fulfilled, this last approach can produce serious
the equation: errors in the quantification.
When a “one-standard addition calibration” is made, the
Y = Y0 + Ym + Sm · (Xnative + Xadded ) quantification step requires the analysis of at least two test por-
= Y0 + Ym + Sm · Xnative + Sm · Xadded tions, one of them containing an amount of analyte standard
added. Then the sample analyte amount is obtained applying
where Xnative indicates the analyte amount in each test portion the next equation which is a particular case of the general one
which is related with the analyte originally present in the sam- above explained:
ple and Xadded represents the analyte amount added to each test
ya,(spl+add) − ya,spl 1
portion. The AC-curve fitted (measured analytical signal versus RF = ; CF =
added analyte amount) is given by: xa,add RF

ya,(spl+added) = aAC + bAC · xa,added ya,spl


xa,spl = CF · ya,spl = xa,add ·
ya,(spl+add) − ya,spl
whose intercept (see Fig. 2a and b) is:
being ya,spl and ya,(spl + add) the analyte signals in the original and
aAC = y0 + ym + Sm · xa,spl
the added samples, respectively; and xa,add is the added standard
where Sm is the AC-slope (bAC ), y0 is the method blank, ym is analyte amount.
the matrix blank, ya,spl and ya,(spl + add) are the analyte signals in As in the previous methodologies, the AC-calibration could
the original and added samples, respectively, and xa,add is the be performed from signals-ratio. The equations for calibration
added analyte amount. and quantification are equivalent to the previous ones without
From this expression is possible to quantify the analyte more than to substitute the analytical signals for the respective
amount initially present in the sample (xa,spl ): analytical signals-ratio.
This type of calibration is the only methodology in which
aAC − (y0 + ym ) aAC − (y0 + ym )
xa,spl = = the results are not affected by systematic matrix errors, inde-
Sm bAC pendently of the kind and magnitude of the matrix effect, so it
and as it can be seen, it is necessary to correct the intercept permits to quantify the amount of analytes in any kind of sam-
value (aAC ) with the total sample blank (method blank + matrix ples. That is why a “in-house” validation of analytical methods
blank). based on the standard additions and Youden calibrations has
On the other hand, in order to obtain a bias free result by been proposed [40]. Its principal disadvantage lies in the fact that
means AC, it is necessary to carry out a “Youden calibration” a calibration for each sample is needed, which implies a lot of
(YC) to estimate the total blank of the sample. It cannot be work in routine analysis. To avoid this problem, a simulation of a
L. Cuadros-Rodrı́guez et al. / J. Chromatogr. A 1158 (2007) 33–46 43

MC-curve by means of an empirical calibration curve obtained method”. Initially, the quantification involved the average geo-
from the YC- and AC-curves features, has been recently pro- metrical value calculation of the two results obtained with each
posed. This new calibration function, named “matrix-corrected standard:
calibration” (MCC ) [41], is a hypothetical matrix-matched cal- 
ibration characterized by a linear curve where the x-variable is xa,spl = ya,spl CFIS1 · CFIS2
the analyte amount (in the presence of matrix) and the y-variable This equation can be significantly simplified when: (i) two
is the analytical signal (Fig. 2b). closely related compounds, upper and lower, of the same
To establish the MCC , is necessary to set up only two cali- homologous series to which the analyte belongs, are used
brations: AC and YC, which must be representative. From the as internal standards (e.g. methanol and n-propanol could be
intercept (aAC ) and the slope (bAC ) of the AC and the intercept used for ethanol determination); (ii) the analyte and inter-
(aYC ) of the YC, both MCC -intercept and MCC -slope can be nal standards calibration factors are approximately equals
calculated by the equations: (CFIS,1 ≤ CFa ≤ CFIS,2 ), being the product of the relative cali-
bration factors analyte/internal standards around 1 (CF1 CF2 = 1)
aMCC = aAC − ya,spl = aYC ; bMCC = bAC
due to its differences are compensated each other [42]; and
which are exclusive of each type of matrix, since each matrix (iii) both internal standards are added in the same amount,
has its own AC-curve because the AC-slope, which corrects the xIS,1 = xIS,2 = xIS . Then, the analyte amount in the sample can
proportional error introduced by the matrix, may change when be calculated according to:
different matrix amounts are used and consequently bMCC = 
1 
bAC and they are only valid provided that the amount of sample xa,spl =ya,spl · xIS,spl =xIS,spl y1,spl
R · y2,spl
R
used for analysis remains constant. yIS1,spl · yIS2,spl
Thus, when a matrix effect exists, the analyte amount from
each measured analytical signal in any sample, can be calculated where ya,1R and yR represent the analyte/internal standard
a,2
from: signal-ratios 1 and 2, respectively, obtained from the sample
analysis.
ya,spl − aMCC ya,spl − aYC The main advantage argued by the authors, arises from that
xa,spl = =
bMCC bAC this quantification method does not need a previous calibration
(the calibration is implicit in the quantification), so its prac-
tical application is very easy. It only requires the addition of
4.4. Internal calibration a known and equal amount of both internal standards to each
sample. Once the sample treatment has been finished, the ana-
There are some analytical situations in which an internal stan- lyte and internal standards signals are measured. The results
dard, employed as calibrant, is added to the sample and analysed obtained show a great accuracy degree, even whether during
as a whole together with the analyte. From the same analytical sample treatment any analyte losing is produced. Its more impor-
preparation, a response/calibration factor is obtained from the tant disadvantage lies in the necessity of having two analyte
IS which is applied on the analyte signal for quantification using homologous (internal standards), initially absents in the sample.
the following equations: To apply an IC the measurement system has to be able to dis-
yIS,spl 1 tinguish between the signals from the internal standard(s) and
RFIS = ; CFIS = the very analyte, that is why, both substances must have a differ-
xIS,spl RFIS
ent behaviour in the measurement system but similar analytical
properties in order to be measured in a quasi-simultaneous way.
ya,spl
xa,spl = CFIS · ya,spl = xIS,spl · = xIS,spl · yspl
R
In the same portion of the test sample, the signal of the standard
yIS,spl and the very analyte are measured; from the first one a calibra-
Metrologically, this is the real internal calibration (IC), a one- tion factor is obtained, which is used to quantify the amount of
standard internal calibration methodology formally different to analyte from the second signal.
the signal-ratio calibration; however this last is the most known An exclusive feature of IC is that it is possible to carry out the
application of the use of internal standard in analytical sciences quantification of several analytes (generally from the same fam-
which sometimes has derived in an inappropriate use of the term ily) in the same sample portion, starting from an only internal
internal calibration. calibration, this is, the internal standard can represent to several
The preceding equations could be combined and the quan- analytes in a simultaneous way in the same sample. As conse-
tification would be obtained from the following simplified one: quence, it would be possible to calculate the (percentage) mass
fraction of each analyte using only the values of the measured
xIS,spl analytical signals by:
xa,spl = · ya,spl = xIS,spl · yspl
R
yIS,spl xa,spl FCIS · ya,spl ya,spl
ca,spl (%) =  · 100 =  · 100 =  · 100
xi (FCIS · yi ) yi
With the aim to enhance the accuracy of the results, a modifi-
cation of the methodology that uses two internal standards, has where ca,spl indicates the percentage mass fraction of the analyte
been proposed and it has been called “double standard internal “a”; xa,spl is the mass of this analyte in the sample and ya,spl is
44 L. Cuadros-Rodrı́guez et al. / J. Chromatogr. A 1158 (2007) 33–46

the analytical signal due to it. xi,spl represents the sum of the the signal decreasing during the analytical sequence when the
masses of all the analytes and, yi,spl represents the sum of the detector response shows a significant instability with the time.
analytical signals attributed to these ones. This strategy is only
possible when the calibration factors are similar for all the ana- 4.5. Calibration by internal normalization
lytes, and equal to the calibration factor of the internal standard.
The reliability of internal calibration as calibration/ When the measuring system sensitivity changes from one
quantification methodology is based on the ability of the IS to analyte to another, the application of the internal standard cali-
act as surrogate. In metrological terms, this fact implies that the bration requires to normalize the measured analytical signals by
response factor (or sensitivity) of the IS is constant and equal determining a “normalization factor” (NF) for each analyte in
to the calibration factors of each one of the substituted analytes, relation to a reference compound (usually, an internal standard).
into the interval of application of the method. That can be true This calibration methodology is called “calibration by internal
for those universal no-specific detectors which respond directly normalization” or “normalized signal calibration”.
to the substance amount (mass or concentration), as it happens Despite its denomination, it is properly a methodology of
with the most classical GC detectors. However, this condition double calibration external–internal, applied in two steps: (i) in
is harder to fulfil in LC, due to the majority of common detec- the first, a multiple external standard measurement to determine
tors used, produce a dissimilar signal amount depending on the the calibration factors of all the components (analytes and inter-
judged analyte. Even, when MS detectors are used, and depend- nal standard) and, the normalization factors of the analytes in
ing on the ionization system, it could be consider this premise relation with the internal standard, are used; and (ii) in the sec-
is satisfied when they work in “full-scan” mode (not in “SIM” ond, the calibration factor of the internal standard, previously
mode). In any case, and as far as possible, it is necessary to verify added to the sample, is determined and the analytes are quanti-
that this requirement is achieved using a measurement standard fied. Thus, the advantages of both methodologies are joined and
containing both analyte and internal standard. the drawbacks of each one of them when are applied separately,
The main advantage of internal calibration is the reduction are minimized.
of analysis time, since it only needs one analytical preparation The NF values of the analytes depend on the detector
for calibration and quantification. Moreover, this methodology employed, so it is necessary to estimate them for each measur-
permits to make up for the losing of analyte during sample prepa- ing system. For this purpose, in the first step, a multiple standard
ration and, in a moderate way, the matrix effect. That is why it is calibration, which includes a known amount of each analyte and
advisable when at least one of the following circumstances are internal standard, must be prepared and measured previously to
produced: (i) the previous preparation sample process is long the sample. Then, the different normalization factors are cal-
and complicated; (ii) a long time for the measure is required culated as the quotient between the calibration factors of the
(e.g. chromatogram run time); and (iii) there is no or it is impos- considered analyte (CFa ) and the internal standard (CFIS ):
sible to acquire a material with analyte standard. Obviously, the
CF xa,std /ya,std xa,std yIS,std xa,std 1
main limitation of this methodology is the availability of an ade- NF = = = · = · R
quate IS. As an additional precautionary measure, the amount of CFIS xIS,std /yIS,std xIS,std ya,std xIS,std ystd
IS must be such its analytical signal (peak area or peak height) Consequently, the NFa,std is the relative calibration factor of each
be of the same order that those corresponding to the analyte; analyte in relation to a reference compound (internal standard).
otherwise, significant errors in quantification can be found, over It can be underlined that this NF equation is similar to the one
all if the internal standard signal is not entirely linear. used to calculate the relative calibration factors (RCF) in Section
UE olive oil official analytical methods for the determina- 4.1.
tion of the composition and content of waxes [43], sterols [44], In the second step, the calculated NF values are used to esti-
stigmastadienes [45], and aliphatic alcohols [46], by capillary- mate a normalized signal for each analyte in the sample (ya,spl N )
column gas chromatography, constitute examples of common from the measured signals (ya,spl ):
application of IC. In those methods, an only calibration IS is
used for the quantification of some analytes of the same chemical ya,spl
N
= NF · ya,spl
family.
An imaginative proposal recently reported is the “echo-peak These normalized signals indicate the hypothetical signals that
internal standard calibration” [47]. This is a novel calibra- would be measured if the calibration factors were similar for
tion methodology which simulates the use of internal standard all of them, and equal to the calibration factor of the internal
by injecting consecutively, within a short period of time, an standard. Finally, the normalized signals can be used for quan-
unknown sample and a standard analyte solution. As a result, tification of the analytes in the sample, by applying the next
the standard analyte peak elutes closely to the sample analyte expression:
peak, thus performing a “echo peak”. For quantification pur-
xa,spl = CFIS · ya,spl
N
= CFIS · NF · ya,spl
poses, a calibration function is obtained from the signal-ratios
of the standard and the reference standard. The concentration of where it can be seen that, the calibration factor of each analyte
analyte in the unknown sample is calculated from the peak area (CF), can be unfolded into two factors:
ratio of sample and reference. As an internal standard method,
the echo-peak technique provide the possibility to compensate CF = NF · CFIS
L. Cuadros-Rodrı́guez et al. / J. Chromatogr. A 1158 (2007) 33–46 45

To assure the validity of this methodology, it is necessary that nal calibration with an only standard (an one-standard external
the calibration factor of internal standard keep constant. calibration) until the most tedious, for routine analysis, standard
As it has been stated in the previous section, is possible to addition calibration from several calibration standards prepared
obtain directly the (percentage) mass fractions of each analyte and measured in replicate.
in a sample containing several analyte from the same chemical The selection of the appropriate methodology should be
family. Whether only the analyte mass fraction in the sample is based on a good knowledge of the metrological features of
required, any other of the analytes in the sample can be used the CMP and/or from a previous judgment on particular situ-
as reference compound to normalize the signals, not being nec- ations such as lack of stability or drift of the measuring system,
essary to add an internal standard neither in the measurement existence of the (additive or proportional) matrix effects, avail-
standard nor in the sample. ability of proper calibration standard (from the analyte or some
In this case, the NF values are calculated directly from the surrogates), etc. The paper also contains a lot of practical
mass fraction (in percentage or non) of each analyte “i” in the recommendations and sufficient guidance to properly apply a
calibration standard (ca,std ) as: chromatographical (or from other separation technique) cali-
 bration, in order to avoid misconceptions and errors for lack of
xi,std yref,std xi,std / xi,std yref,std knowledge.
NF = · =  ·
xref,std yi,std xref,std / xi,std yi,std
ci,std yref,std
= · Acknowledgement
cref,std yi,std
where the subscript “ref” is referred to the analyte reference, We acknowledge financial support from the Spanish Min-

and the term xi,std is the sum of the amounts of all analyte, isterio de Educación y Ciencia (MEC), Dirección General de
including the analyte reference. Thus, the composition of one Investigación (Project No. CTQ2006-15066-C02-02).
analyte “a” in the sample (ca,spl ), expressed as mass fraction (in
percentage), is calculated as:
References
xa,spl CFref · ya,spl
N
ca,spl (%) =  · 100 =  · 100 [1] Nomenclature in Evaluation of Analytical Methods including Detection
xi,spl (CFref · yi,spl
N )
and Quantification Capabilities, IUPAC Recommendations 1995 (prepared
by L.A. Currie), Pure Appl. Chem. 67 (1995) 1699–1723; reproduced in
ya,spl
N
NFa · ya,spl L. A. Currie, Anal. Chim. Acta 391 (1999) 105.
=  N · 100 =  · 100 [2] International Vocabulary of Metrology—Basic and General Concepts and
yi,spl (NFi · yi,spl ) Associated Terms (VIM), third ed., Final Draft, August 2006, http://www.
ncsli.org/vim/wg2 doc N318 VIM 3rd edition 2006-08-01%20(3).pdf.
The calibration by internal normalization is a methodology [3] L. Cuadros Rodrı́guez, L. Gámiz Gracia, E. Almansa López, J. Laso
which has been scarcely used in analytical quantification. How- Sánchez, Trends Anal. Chem. 20 (2001) 195.
ever, it exits a significant application example in the UE official [4] A. Switaj-Zawadka, P. Konieczka, E. Przyk, J. Namiesnik, Anal. Lett. 38
analytical method for the determination of methyl esters of fatty (2005) 353.
[5] International Vocabulary of Basic and General Terms in Metrology (VIM),
acids by gas chromatography in olive oil [48], a representative second ed., International Organization for Standardization (ISO), Geneva,
example in which both ones, an internal standard and an analyte, 1993.
are used as reference compound. [6] ISO 11843-1, Capability of Detection—Part 1: Terms and Definitions,
International Organization for Standardization (ISO), Geneva, 1997.
[7] Guidelines for Calibration in Analytical Chemistry—Part 1: Fundamen-
5. Conclusion tals and Single Component Calibration, IUPAC Recommendation 1998
(prepared by K. Danzer and L.A. Currie), Pure Appl. Chem. 70 (1998)
An overview on the problem of the analytical calibra- 993.
tion/qualification for the separation sciences has been shown up. [8] Harmonized Guidelines for Internal Quality Control in Analytical Chem-
The paper intends to establish a description of the metrological istry Laboratories, IUPAC Technical Report 1995 (prepared by M.
Thompson and R. Wood), Pure Appl. Chem. 67 (1995) 649.
fundamentals with the aim of reconciling the vocabulary and the [9] Absolute Methods in Analytical Chemistry, IUPAC Technical Report 1995
subjects of the measurement science with those corresponding (prepared by A. Hulanicki), Pure Appl. Chem. 67 (1995) 1905.
to the chemical analysis, when separation techniques are previ- [10] Selectivity in Analytical Chemistry, IUPAC Recommendations 2001 (pre-
ously applied. We believe necessary that the chromatographers pared by J. Vessman and col.), Pure Appl. Chem. 73 (2001) 1381.
familiarize with this terminology that nowadays is considered [11] K. Danzer, Anal. Bioanal. Chem. 380 (2004) 376.
[12] Quality Control Procedures for Pesticide Residue Analysis, Document
as a fundamental topic of the analytical chemistry. SANCO/10232/2006, European Union, Brussels, 2006.
In addition, on the basis of the conventional multi-standard [13] L. Cuadros Rodrı́guez, L. Gámiz Gracia, E. Almansa López, J.M. Bosque
calibration, the principles, advantages and limitations of alter- Sendra, Trends Anal. Chem. 20 (2001) 620.
native calibration schemes as two- and one-standard calibration [14] ISO 8466, Water Quality—Calibration and Evaluation of Analytical Meth-
have been presented. The different calibration methodologies, ods and Estimation of Performance Characteristics (Parts 1 and 2),
International Organization for Standardization (ISO), Geneva, 1990 and
that could be applied, have been explained and their scope have 1993.
been discussed. We have tried to show how the analysts can have [15] ISO 11095, Linear Calibration using Reference Materials, International
a battery of possibilities that embraces from the simplest exter- Organization for Standardization (ISO), Geneva, 1992.
46 L. Cuadros-Rodrı́guez et al. / J. Chromatogr. A 1158 (2007) 33–46

[16] LGC/VAM/2003/032, Preparation of Calibration Curves—A Guide to Best [33] M.J. Cardone, J. AOAC Int. 66 (1983) 1283.
Practice, Laboratory of the Government Chemist (LGC), Cambridge, 2003. [34] Guidelines, for Single-Laboratory Validation of Analytical Methods
[17] J.N. Miller, Analyst 116 (1991) 3. for Trace-Level Concentration of Organic Chemicals, Association of
[18] D.L. MacTaggar, S.O. Farwell, J. AOAC Int. 75 (1992) 594. Official Analytical Chemist (AOAC)/Food and Agriculture Organiza-
[19] K. Baumann, H. Waetzig, Proc. Cont. Qual. 10 (1997) 59. tion of the United Nations (FAO)/International Atomic Energy Agency
[20] K. Baumann, Proc. Cont. Qual. 10 (1997) 75. (IAEA)/International Union of Pure and Applied Chemistry (IUPAC).
[21] R. de Levie, Crit. Rev. Anal. Chem. 30 (2000) 59. [35] A.R.C. Hill, S.L. Reynolds, Analyst 124 (1999) 953.
[22] E.A.-04/14, The Selection and Use of Reference Materials, European Co- [36] J.L. Martı́nez Vidal, F.J. Arrebola, A. Garrido Frenich, J. Martı́nez
operation for Accreditation, 2003. Fernández, M. Mateu Sánchez, Chromatographia 59 (2004) 321.
[23] Harmonized Guidelines for Single-Laboratory Validation of Methods of [37] J.L. Martı́nez Vidal, A. Garrido Frenich, T. López López, I. Martı́nez Sal-
Analysis, IUPAC Technical Report 2002 (prepared by M. Thompson, vador, L. Hajjaj el Hassani, M. Hassan Benajiba, Chromatographia 61
S.L.R. Ellison and R. Wood), Pure Appl. Chem. 74 (2002) 835. (2005) 127.
[24] A. González Casado, A.M. Garcı́a Campaña, L. Cuadros Rodrı́guez, J.L. [38] P. Koscielniak, J. Kozak, Crit. Rev. Anal. Chem. 36 (2006) 27.
Vilchez Quero, R. Blanc Garcı́a, LC GC Int. 11 (1998) 726. [39] R.C. Castell, M.A. Castillo, Anal. Chim. Acta 423 (2000) 179.
[25] K. Altria, LC GC Eur. 1 (September) (2002) 1. [40] A.M. Garcı́a Campaña, L. Cuadros Rodrı́guez, J. Aybar Muñoz, F. Alés
[26] L. Cuadros Rodrı́guez, A. González Casado, A.M. Garcı́a Campaña, J.L. Barrero, J. AOAC Int. 80 (1997) 657.
Vı́lchez, Chromatographia 47 (1998) 550. [41] I. Fernández Fı́gares, A. González Casado, L. Cuadros Rodrı́guez, J. Chro-
[27] A.M. Gómez Caravaca, A. Carrasco Pancorbo, B. Cañabate Dı́az, A. matogr. B 779 (2004) 73.
Segura Carretero, A. Fernández Gutiérrez, Electrophoresis 26 (2005) [42] I.G. Zenkevich, E.D. Makarov, J. Chromatogr. A, in press.
3538. [43] Commission Regulation (EC) No. 183/93, Annex IV, Off. J. Eur. Commun.
[28] Method 1613, US Environmental Protection Agency (EPA), Washington, L22 (1993) 62.
DC, 1994, http://www.epa.gov/waterscience/methods/1613.pdf. [44] Commission Regulation (EC) No. 2568/91, Annex V, Off. J. Eur. Commun.
[29] Methods 7.2 and 7.3, Harmonised Methods of the International Honey L248 (1991) 15.
Commission (IHC), Berne, 2002, http://www.apis.admin.ch/enghish/host/ [45] Commission Regulation (EC) No. 656/95, Annex XVII, Off. J. Eur. Com-
pdf/honey/ICHmethods e.pdf. mun. L69 (1995) 05.
[30] ISO Guide 32, Use of Certified Reference Materials, International Organi- [46] Commission Regulation (EC) No. 769/2002, Annex XIX, Off. J. Eur. Com-
zation for Standardization (ISO), Geneva, 1997. mun. L128 (2002) 23.
[31] M. Thompson, S.L.R. Ellison, Accred. Qual. Assur. 10 (2005) 82. [47] J. Zrostliková, J. Hajslová, J. Poustka, P. Begany, J. Chromatogr. A 973
[32] L. Cuadros Rodrı́guez, A.M. Garcı́a Campaña, E.M. Almansa López, F.J. (2002) 13.
Egea González, M.L. Castro Cano, A. Garrido Frenich, J.L. Martı́nez Vidal, [48] Commission Regulation (EC) No. 2568/91, Annex X-A, Off. J. Eur. Com-
Anal. Chim. Acta 478 (2003) 281. mun. L248 (1991) 36.

View publication stats

You might also like