J Solid State Electrochem (2011) 15:6976

DOI 10.1007/s10008-010-1106-9

REVIEW

The history and development of a rigorous metrological basis

for pH measurements
Petra Spitzer & Kenneth W. Pratt

Received: 8 March 2010 / Revised: 3 May 2010 / Accepted: 10 May 2010 / Published online: 4 June 2010
# Springer-Verlag 2010

Abstract This paper discusses the basis and historical

development of the traceability chain for pH. The quantity
pH, first introduced in 1909, is among the most frequently
measured analytical quantities. The practical measurement
of the pH value of a sample is inexpensive, easy to perform,
and yields a rapid result. However, the problems posed by
the traceability of pH are not easy to solve. Most pH
measurements are performed by potentiometry, using a
glass electrode as the pH sensor. Such pH electrodes must
be calibrated at regular intervals. Confidence in the
reliability of pH measurements requires establishment of a
metrological hierarchy including an uncertainty budget for
calibration that links the pH measured in the sample to an
internationally agreed and stated reference. For pH, this
reference is the primary measurement of pH. A traceability
chain can be established that links field measurements of
pH to primary buffer solutions that are certified using this
primary method. This allows the user in the field to
estimate the measurement uncertainty of the measured pH
data. As the realization of the primary measurement is
sophisticated and time-consuming, primary standards are
generally realized at national metrology institutes. A
number of potentiometric methods are suitable for the
determination of the pH of reference buffer solutions by
comparison with the primary standard buffers. The choice

P. Spitzer (*)
Physikalisch-Technische Bundesanastalt (PTB),
Bundesallee 100,
38116 Braunschweig, Germany
e-mail: [email protected]
K. W. Pratt
National Institute of Standards and Technology (NIST),
100 Bureau Dr., Stop 8391,
Gaithersburg, MD 20899-8391, USA

between the methods should be made according to the

uncertainty required for the application. For reference
buffer solutions that have the same nominal composition
as the primary standard, the differential potentiometric cell,
often called the Baucke cell, is recommended.
Keywords pH . Potentiometry . Metrology . Traceability .
Reference standard . Primary measurement . Differential cell

Introduction
The quantity pH is a measure of the acidity of an aqueous
solution. Many chemical and physiological processes are
affected by the pH. Since the solubility and bioavailability
of substances are each functions of the pH, it is also a
critical input quantity for climate modeling. The rate of
chemical reactions can be altered by pH. Research in
environmental science and in biochemistry has given
evidence that the measurement uncertainty in pH forms a
central contribution to the uncertainty budget of thermodynamic data as well as of geochemical transport models [1].
The first survey article on pH, published in 1914 [2],
mentioned the importance of the hydrogen ion concentration and its measurement in biology.
More than 100 years ago, S. P. L. Srensen introduced
the notation pH as a practical expression derived from the
measurement of the hydrogen ion concentration [3].
Today, pH measurements are carried out on a large scale,
both at the laboratory level and in industry. In fact, pH is
among the most frequently measured analytical quantities.
The measurement of the pH value of a sample is
inexpensive easy to perform and yields a rapid result. In
spite of these advantages, the problems posed by the
traceability of pH are not easy to solve. The prerequisite

70

for mutual acceptance of analytical data is comparability of

the results, even if these data are measured by different
analysts, using different equipment, and at different times
and places [4]. Statements of the measurement uncertainty
based on traceability to internationally recognized references are required by regulatory bodies or by international
quality standards in general [5].
Most pH measurements are performed by potentiometry.
In almost all cases, a glass electrode is the indicating pH
sensor. A pH electrode used in the field must be calibrated
at regular intervals.
Confidence in the reliability of measurement results
requires establishment of a metrological calibration hierarchy
linking the quantity in the sample to a unit in the International
System of Units (SI) or, where this is not possible, to another
international agreed and stated reference.
For pH, the result of a calibration must be compared
with the value provided by a certified reference buffer
solution. The pH value of the reference buffer solution and
its uncertainty are assigned using a measurement procedure
that itself also must be calibrated. In this way, a traceability
chain can be established linking the pH measured in the
field to the pH of a primary buffer solution. Primary
measurement standards and primary measurement procedures are typically realized by National Metrology Institutes (NMI) in their respective countries.
In the following, the basis and the historical development of the traceability chain for pH is described.

The concept of pH
In the last decade of the nineteenth century, Arrhenius
introduced the ionic theory [6]. This theory provides the
fundamental relationship between hydrogen and hydroxyl
ions and offers the possibility to determine the dissociation
constant of water.
In 1904, Friedenthal prepared 14 solutions of known
hydrogen concentration and examined the color change of
indicator dyes corresponding to a particular hydrogen ion
concentration. Friedenthal for the first time established a
scale of acidity [7].
In 1909, Srensen, a coworker of the Carlsberg
Laboratory in Copenhagen, defined pH in terms of the
concentration with a scale of 014 (at 25 C) derived from
the ionic product of water [3]. In the same year, the glass
electrode was applied to acidimetric titrations for the first
time by Haber and Klemensiewicz [8].
Srensen's original intention was to simplify the inconvenient figures that result from measurements of the
hydrogen ion concentration. Srensen proposed to take
the decadic logarithm of the hydrogen ion concentration
multiplied by (1) and called it pondus Hydrogenii.

J Solid State Electrochem (2011) 15:6976

Some years later, the concept of activity was introduced [9],

and in 1923, Debye and Hckel published [10] the theory
of interionic interaction. On the basis of this knowledge,
Srensen and K. Linderstrm-Lang [11] proposed the
definition of pH generally accepted today, which is based
on the activity of the hydrogen ions in solution:


pH  log aH  log mH g H =m0 :

In Eq. 1, aH is the molality-based activity; H, the molal

activity coefficient of the hydrogen ion H+ at the molality
mH; and mo is a standard state, chosen to be equal to
1 mol kg-1.
The single ion activity involved in Eq. 1 cannot be
measured unambiguously. It is necessary to introduce a
convention based on approximations that are related to
measurable quantities, to approach the definition of pH as
closely as possible.
Since its introduction by Srensen, the interpretation of
the quantity pH has given rise to intense discussion and
controversies. Bates, in 1948, stated that several scales, all
masquerading under the name pH, are in common use.
Many investigators are thinking and computing in terms of
one definition and measuring a different quantity [12].
Several pH scales have been described which are
characterized by standard buffer solutions with assigned
pH values [13].
In the 1960s, Bates and co-workers at the National
Bureau of Standards (NBS), now the National Institute of
Standards and Technology, established the so-called multipoint pH protocol after extensive studies of buffer solutions
and suitable electrochemical cells [14]. In the multipoint pH
protocol, each pH standard is a dilute aqueous solution with
an ionic strength, I0.1 mol/kg. The pH values of these
standards are measured using a cell without liquid junction,
the so-called Harned cell [15, 16], described below in
further detail. This protocol was recommended in 1979 by
the International Union of Pure and Applied Chemistry
(IUPAC) [17].
In addition to the Harned cell, the 1979 IUPAC
Recommendation also presented an operational definition
of pH. The pH of an unknown solution, pH(X), in a cell
containing hydrogen and a reference electrode with a liquid
junction was related to the pH of one standard solution, pH
(S), by Eq. 2:
pHX pHS

E S  E XF
:
RT ln 10

In Eq. 2, the quantities E(S) and E(X) are the potentials

of this cell containing the standard solution and the
unknown sample; and pH(X) and pH(S) denote the pH of
the unknown solution and the standard, respectively. The
quantities R, T, and F represent the gas constant; thermo-

J Solid State Electrochem (2011) 15:6976

dynamic temperature, and the Faraday constant, respectively.

The quantity RT ln10/F is often referred to as the
thermodynamic or Nernstian slope, k. Equation 2 has a
major shortcoming: it excludes the contribution from the
liquid junction potential (LJP). The LJP arises at the liquid
junction between the reference electrode and the solutions
S or X and is a complex function of the solution
composition and other factors. The difference in the
residual LJP between solutions S and X is a fundamental
source of bias in the operational measurement of pH. Thus
defined, the quantity pH is a number. It was allowed, to a
good approximation, to replace the hydrogen electrodes
in both cells by other hydrogen-ion-responsive electrodes,
e.g., glass or quinhydrone [17].
In 1981, the British Standard Organization (BSI)
recommended [18] a so-called single-standard scale. The
pH of a single reference standard, a solution of potassium
hydrogen phthalate with a molality of 0.05 mol/kg, was
measured in a cell without liquid junction. The resulting pH
was defined as a fixed reference point. The pH values of
other solutions in the BSI protocol were defined by the
operational definition as realized in a cell with a specified
geometry of the liquid junction.
These two definitions of pH coexisted for a time.
During this period, all attempts to harmonize the pH scale
recommendations [19] were without success.
In 1985, the IUPAC issued a recommendation, Definition
of pH scales, standard Reference values, measurement of pH
and related terminology [20]. This proposal was edited as a
compromise recommendation and recognized the validity
of both the NBS and BSI approaches. This situation was
highly unsatisfactory. The application of two pH scales in
parallel resulted in two possible pH values for each unknown
solution.
The 1985 IUPAC Recommendation was seen as an
interim document from the beginning. The recommendation
of two incompatible protocols for pH was to be replaced by
amended recommendations when a thermodynamically
and metrologically sound pH scale can be recommended
[21]. In the 1990s, pH measurement was one of the first
topics addressed in the application of metrological principles to chemistry [22]. The issuance of a new IUPAC
document became a necessity, owing to several additional
metrological shortcomings of the 1985 IUPAC Recommendation. Notably, the 1985 Recommendation did not address
the uncertainty of pH measurements, a topic of increasing
importance with the introduction of metrological principles
to the field. Also, various different pH definitions and
recommendations to measure pH had been uncovered by a
critical analysis of the document by F. G. K. Baucke [23].
Baucke, at that time Chairman of the DIN Committee
Technical pH measurement was one of the initiators of a
Working Party on pH which was established in 1997 [24].

71

The task for this Working Party was to prepare amended

recommendations to replace the 1985 document. After
extended deliberations, in 2002, the IUPAC issued a new
Recommendation for a Primary Measurement Procedure
for pH [25]. This document eliminated the dual pH scale
of the 1985 document and today forms the basis for the
primary measurement of pH.

The primary method for the measurement of pH

As presented, in the 2002 Recommendation, the primary
procedure for pH is based on the measurement of the
potential difference of an electrochemical cell without
liquid junction, involving a selected buffer solution, a
platinum hydrogen gas electrode, and a silver/silver
chloride reference electrode, in cell I [26]:
PtjH2 jbuffer; Cl jAgCljAg:

Cell I

Cell I was originally developed by Harned and coworkers in the 1930s [15] and is frequently called the
Harned cell. Harned originally applied cell I to the
determination of activity coefficients. Subsequently, Hamer
and Acree at the NBS suggested the use of cell I in accurate
pH measurements [27]. This work was developed further at
the NBS by Bates [28] and by Hamer and Acree [29]
starting in the 1940s.
The principle of cell I is shown in Fig. 1.
Chloride ions are added to the chloride-free buffer at
several chloride molalities in order to stabilize the potential
of the silver/silver chloride electrode. The Harned cell
avoids the liquid junction potential and its associated
difficulties. The cell potential consists merely of the
difference in the two electrode potentials.
In addition to the use of a cell without liquid junction, a
convention is necessary to obtain the single ion activity
embodied in Eq. 1 above. The primary method for pH uses

Fig. 1 Platinum, hydrogen | silver, silver chloride cell (so-called

Harned cell) for primary pH measurements

72

J Solid State Electrochem (2011) 15:6976

the BatesGuggenheim convention [30]. Bates and

Guggenheim suggested this approximation for the single
ion activity coefficient based on the DebyeHckel theory
of strong electrolytes. This convention is restricted to
solutions of low ionic strength, I0.1 mol kg-1.
The potential difference E of cell I (corrected to
101325 Pa partial pressure of hydrogen) depends on the
hydrogen ion activity, aH, according to Eq. 3:




E E 0  k log aH =m0 mCl g Cl =m0 :
3
In Eq. 3, E0 denotes the standard potential of the Ag/AgCl
reference electrode, mCl and Cl the molality and activity
coefficient of the chloride ion. Equation 3 is the well-known
Nernst equation applied to the Harned cell.
A pH measurement of a buffer solution using cell I usually
consists of different steps [31]. In the first step, E0 of the
silver/silver chloride electrode is determined in a Harned cell
filled with hydrochloric acid of fixed molality. The mean
activity coefficient of HCl, g HCl , at various temperatures is
best known at the molality 0.01 mol kg-1 [32].
In the second step, the Harned cell is filled with the
buffer solution whose pH value is to be measured. A new
quantity, the acidity function, pa, is defined by Eq. 4:


pa  log aH g Cl =m0 :
4
Equations 3 and 4 can be combined to yield Eq. 5, in
which pa is a function only of measurable quantities:




pa E  E0 =k log mCl =m0 :
5
In the primary procedure for pH, pa is measured as a
function of mCl.
The third step is the extrapolation of the acidity function to
zero chloride molality. This extrapolation yields the acidity
function corresponding to zero chloride molality, pa0:


pa0  log aH g Cl =m0 mCl!0 :
6
The value of pa0 is determined by linear extrapolation of
pa as a function of the chloride molality, using measurements at least three values of mCl in the range from 0.005 to
0.02 molkg-1. It is assumed that a linear extrapolation is
appropriate if the change in ionic strength produced by the
addition of chloride is restricted to less than 20%.
This extrapolation is mathematically depicted in Eq. 7,
where b is an empirical, temperature-dependent constant:
pa pa0 bmCl :

In the fourth step, the activity coefficient at the ionic strength

I of the buffer, C1, is obtained by adopting the Bates
Guggenheim convention. The value of C1 is given by Eq. 8:

1=2 

1=2 
= 1 1:5 I=m0
log g Cl A I=m0
:
8

In Eq. 8, A is the DebyeHckel temperature-dependent

limiting slope [33].
Finally, the pH value of the primary pH buffer solution,
pH(S), is calculated by combining Eqs. 6, 7, 8, and the
definition of the pH, Eq. 1, to yield Eq. 9.
pHS pa0 log g Cl :

Traceability of the measured pH value to the SI can be

established for pH(S) values determined as described above if
the expanded uncertainty associated with the extrathermodynamic assumption, the BatesGuggenheim convention, is taken into account. Unfortunately, the
measurement uncertainty contribution arising from the
use of this convention is 0.01 (expanded measurement
uncertainty, coverage factor k=2). In comparison, the
experimental expanded measurement uncertainty (k=2)
obtained for a pH value of a typical primary buffer
solution is 0.003 to 0.004. Values of pH that include all
sources of uncertainty excepting that of the Bates
Guggenheim convention, as is the common practice at
most NMIs, are considered conventional pH values, which
is sufficient for most applications [21].

Primary pH reference buffers

The primary buffer solutions currently recommended by
IUPAC establish the so-called multipoint pH protocol [14].
The primary method for pH is used by the NMIs to assign
pH values to a restricted number of primary reference
solutions in dilute aqueous solutions between pH 3 and 10
and in a temperature range from 5 C to 50 C. Typical
values of the pH of primary reference buffer solutions as
listed in the 2002 IUPAC Recommendation [25] and also in
the ISO [34] are listed in Table 1.
Several criteria must be fulfilled for a material to qualify
as a primary pH buffer material. Primary buffer materials
must be available at high purity, have high stability, and
good reproducibility of preparation. The buffer solutions
themselves have been selected such that only small liquid
junction potentials occur in measurements made with
practical pH electrodes incorporating liquid junctions. Of
course, the magnitude of the residual liquid junction
potential also depends on the kind of liquid junction device
(diaphragm) used.
Each batch of material must be individually certified.
The pH and the associated uncertainty are given in the
certificate for each measurement temperature. Batch to
batch variations are of the same order as the expanded
measurement uncertainty of the primary method.
Calcium hydroxide and potassium tetraoxalate are also
certified using the above primary measurement procedure.

J Solid State Electrochem (2011) 15:6976

73

Table 1 Typical values of pH for primary reference buffer at 25 C

Primary pH reference buffer

Molality in mol kg-1

Potassium hydrogen tartrate

Potassium hydrogen phthalate
Potassium dihydrogen phosphate/
disodium hydrogen phosphate

Saturated at 25 C
0.05
0.025/0.025
0.008695/0.03043
0.01
0.025/0.025

Sodium tetraborate decahydrate

Sodium bicarbonate/sodium carbonate

pH, typical values at 25 C [25]

3.557
4.005
6.865
7.416
9.180
10.012

These figures should not be used in place of the certified value for a specific batch of buffer material

They are also listed, e.g., in the DIN 19266 [35]. However,
they are not recommended by IUPAC as primary buffers,
because the contribution of the hydroxyl or hydrogen ion to
the ionic strength is significant.

The Differential pH Cell (Baucke Cell)

Operating the Harned cell is very sophisticated and timeconsuming. In general, primary measurement standards are
realized by NMIs. In many cases, these standards are not
economically suitable for calibration and reference laboratories or for routine measurements. There are a number of
cells having liquid junctions which may be used for the
determination of the pH of a secondary reference buffer
solution by comparison with a primary buffer solution [36].
Standardization of a secondary buffer solution (often by a
non-NMI entity) extends traceability to a greater market
and offers significant economic advantages.
For reference buffer solutions with the same nominal
composition as that of the primary standard, the differential
potentiometric cell is recommended. Baucke was the first to
apply [3739] a differential cell, cell II, to the standardization of pH buffers:
Pt1jH2 j buffer S1 jjbufferS2 jH2 jPt2:

Cell II

Equation 10 may be rearranged and expressed in terms

of pH:
pHS2 pHS1 

ECell II  Ej
;
k

The substitutions in Eq. 11 of pH(S1) and pH(S2) forE

(1) andE(2) are based on the thermodynamic response of
the Pt|H2 electrode (Nernst equation) to aH and on the
notional definition of pH (pH=log aH) [25].
Provided that S1 and S2 are quasi-identical in composition, |pH(S2)pH(S1)|0.02, and 3<pH<11, measurements
[39] indicate that Eq. 12 holds:
 
Ej  < 0:1jECell II j:
12
In other words, if these three conditions are fulfilled, the
bias associated with Ej in cell II is less than 10% of ECell II,
and pH(S2) may be calculated using Eq. 13 (i.e., neglecting
Ej):
pHS2 pHS1 

ECellX
:
k

10

In Eq. 10, E(1) and E(2) denote the potentials of Pt(1)

and Pt(2), respectively. Ej is the liquid junction potential
that forms between S1 and S2 at the diaphragm. The
correction of E(1) and E(2) to standard pressure cancels in
Eq. 10, as pH2 at Pt(1) equals pH2 at Pt(2).

13

The uncertainty associated with the use of Eq. 13 may be

estimated from Eq. 12.
Cell II is used to assign pH(S2) for a certified reference
material (CRM) as follows. S1 is a primary pH buffer, with
pH(S1) having been determined previously using a Harned

Cell II consists of two identical Pt|H2 electrodes, Pt(1)

and Pt(2); and two quasi-identical buffers, S1 and S2, with
pH values, pH(S1) and pH(S2). A diaphragm, ||, separates
S1 and S2. The cell is constructed such that the H2 pressure,
pH2, at Pt(1) and Pt(2) is identical. The cell is schematically
shown in Fig. 2.
The potential, Ecell II, of cell II is given by Eq. 10:
ECell X E2  E1 Ej :

11

Fig. 2 Differential potentiometric cell (Baucke cell)

74

cell measurement [25]. S2 is a CRM buffer, for which pH

(S2) is to be assigned. Typically, S1 is a small batch
reserved as a primary reference only, and S2 is a large batch
with the same nominal composition as S1. The source
reagent(s) for S2 and those for S1 usually derive from
different lots. The traceability of pH(S2) for the CRM to the
SI is via the differential measurement from pH(S2) to pH
(S1) and the Harned cell measurement from pH(S1) to the
SI. The contribution associated with Eq. 12 to the overall
uncertainty is negligible.
Cell II is frequently referred to as the Baucke cell. Owing
to the simplicity of the measurement as compared with the
primary measurement of pH (Harned cell), the Baucke cell has
been adopted by a number of countries [40, 41] in their
scheme for the dissemination of pH measurement capability.

Routine pH measurements: the glass electrode

Owing to their complexity and restriction to specific
buffers, the primary and differential methods described
above are used at NMIs and calibration laboratories to
realize and disseminate the quantity pH.
This section briefly outlines the methodology for routine pH
measurements, including contributions of Baucke to this field.
Routine pH measurements are generally performed using
a pH measuring setup composed of a pH meter and a pH
electrode. The pH electrode, often misleadingly called a
glass electrode, in fact comprises two separate electrodes:
the glass electrode proper and an external reference

Fig. 3 Metrological traceability

scheme for pH according to
[42]. Uc denotes the expanded
measurement uncertainty
(coverage factor k=2) assigned
to the pH values

J Solid State Electrochem (2011) 15:6976

electrode with a liquid junction (diaphragm). The electrode

pair is generally designed as a combination or single rod. A
typical pH electrode can be represented by Cell III:
AgjAgCljKCljjsample; bufferjglassjinternal bufferpH 7
KCljAgCljAg

Cell III

The Ag|AgCl reference electrode and the KCl solution

(3.0 molL-1 to 3.5 molL-1) on the left side of cell III
comprise the external reference electrode. The diaphragm ||
between the KCl solution and the sample buffer provides
electrical contact while minimizing flow of the KCl
solution into the sample buffer. An LJP arises at this
boundary of solution phases.
Owing to various random and systematic effects, notably
including the LJP, cell III must be calibrated with respect to
reference buffer solutions traceable to primary pH standards.
The calibration procedures can be subdivided into
single-point, two-point, and multipoint calibrations [25].
In each of these calibration procedures, the uncertainties of
the buffer pH values (see preceding sections) and the
residual liquid junction potential must be included in the
calculation of the uncertainty of the pH of the sample.
Single-point calibration uses one standard. The calibration
function is assumed to be a straight line defined by the
measured potential of the standard and the thermodynamic
(Nernstian) slope, k. Single-point calibration is of primary
interest to obtain an approximate pH value prior to a more
precise measurement using two-point or multipoint calibration. The two-point or bracketing procedure is used in most

J Solid State Electrochem (2011) 15:6976

routine pH measurements. This calibration uses two standards with values that bracket the range in which the
unknown lies.
Multipoint calibration [25, 42] is recommended when
minimum uncertainty and maximum consistency are required over a wide range of pH(X) values. The pH values of
reference buffer solutions which differ in composition and
buffer capacity from the primary standard buffers are also
determined by a multipoint calibration. These pH reference
buffers often called technical or ready-to-use buffer. The
calibration of the electrode is calculated from linear
regression of the difference in the cell potential at each of
the calibrant pH values from the least-squares line.
In both the two-point and multipoint calibration protocols, the glass electrode typically yields a practical slope, k
that is slightly smaller than k. The main factor in this
deviation is variation in the LJP with pH [25]. However, by
comparing the response of glass and Pt|H2 electrodes,
Baucke et al. [43] demonstrated that a portion of this
deviation to all appearances results from changes in the
surface activity of silanol groups on the surface of the glass
membrane and is, thus, an inherent property of the glass
electrode itself.

75

the pH of the sample is therefore in the order of 0.02 for

most applications.
The work on the traceability and dissemination of pH is
not yet complete. The BatesGuggenheim convention is
only valid at ionic strengths up to 0.1 mol kg-1. For
applications in clinical chemistry and in environmental
samples (e.g., rainwater, seawater), pH reference buffer
solutions with ionic strengths more similar to these samples
are expected to improve the comparability of measurement
results in these matrices. Further investigations into solution
theory and into the concept of single ion activity are
necessary to overcome present limitations for the primary
procedure for measurement of pH.
Finally, the dissemination of traceability from the NMI
to the user in the field is also an area of activity. A new
IUPAC project: Comparable pH measurements by metrological traceability for improving the scientific basis and
broadening the applicability of pH measurement [47] is
developing applicable recommendations. Project 843 of
EURAMET (European Association of National Metrology
Institutes) gives recommendations for the calibration and
evaluation of pH on site measuring instruments [48].
These activities all continue to build on work originally
conceived and implemented by F.G.K. Baucke.

Conclusion
References
Established traceability structures are the prerequisite for
the global comparability of measurement results. Also due
to the persistent work of F.G.K. Baucke within the IUPAC
Working Party on pH, a metrologically well-founded
primary method for pH is now widely adopted in national
[44] and international [45] standards. The use of a simple
differential potentiometric cell to derive secondary pH
buffer solutions with the same chemical composition as
the corresponding primary buffer, as conceived by Baucke,
has been implemented at many calibration laboratories, as
well as at several NMIs in developing countries.
Figure 3 depicts a traceability scheme for pH according
to [46]. The expanded uncertainty Uc is obtained by
multiplying the standard uncertainty by the coverage factor
k=2. Normally, the pH value lies with a probability of
approximately 95% within the assigned interval of values
[49]. At the secondary level, using the differential pH cell
for calibration, the expanded uncertainty of the resulting pH
is in the same order as for the primary method. These
reference buffer solutions are used in general to determine
pH values of buffer solutions with composition different
from the primary pH buffers with a pH glass electrode. An
expanded uncertainty of 0.01 is assigned to these pH
values. The difference in uncertainty is caused by the
contribution of the residual liquid junction potential to the
overall uncertainty. The expanded uncertainty assigned to

1. Meinrath G, Spitzer P (2000) Microchimica acta 135:155168

2. Michaelis L (1914) Die Wasserstoffionenkonzentration. Ihre
Bedeutung in der Biologie und die Methoden ihrer Messung.
Springer, Berlin
3. Srensen SPL (1909) Biochemische Zeitschrift 21:131
200;201304; continuation (1909) Compt. Rend. Trav. Lab.
Carlsberg 8:1153
4. Brown RJC, Milton MJT (2007) Chem Soc Rev 36:904913.
doi:10.1039/b507452p
5. ISO/IEC 17025:2005 (2005) General requirements for the
competence of testing and calibration laboratories. ISO, Geneva,
Switzerland
6. Arrhenius S (1887) Z Phys Chem 1:631648
7. Friedenthal H (1904) Z Elektrochemie 10:113119
8. Haber F, Klemensiewicz Z (1909) Z Phys Chem 67:385431
9. Lewis GN, Randall M (1923) Thermodynamics and the free
energy of chemical substances. McGraw-Hill, New York
10. Debye P, Hckel E (1923) Phys Z 24:185206
11. Srensen SPL, Linderstrm-Lang K (1924) Compt Rend Trav Lab
Carlsberg 15:140
12. Bates RG (1948) Chem Rev 42:160, cited in Bates RG (1981)
CRC Crit Rev Anal Chem 10:247278
13. Galster H (1991) pH measurement. VCH, Weinheim, pp 4752
14. Bates RG (1973) Determination of pH. Theory and practice, 2nd
edn. Wiley, New York
15. Harned HS, Robinson RA (1928) J Am Chem Soc 50:31573178
16. Harned HS, Ehlers RW (1932) J Am Chem Soc 54:13501357
17. Manual of symbols and terminology for physicochemical quantities and units, 2nd ed., prepared for publication by Whiffen D H
(1979). Pure Appl Chem 51:3031
18. Covington AK (1981) Anal Chim Acta 127:121

76
19. Bates RG (1981) The modern meaning of pH. CRC Crit Rev Anal
Chem 10:247278
20. Covington AK, Bates RG, Durst RA (1985) Pure Appl Chem
57:531542
21. Baucke FGK (2002) Anal Bioanal Chem 374:772777
22. Spitzer P, Eberhardt R, Schmidt I, Sudmeier U (1996) Fresenius J
Anal Chem 356:178181
23. Baucke FGK (1997) In: Spitzer P (ed) PTB-report W-68.
Physikalisch-Technische Bundesanstalt, Braunschweig, pp 1020
24. Baucke FGK, Spitzer P, Naumann R (1998) Anal Chem 70
(7):227A
25. Buck RP, Rondinini S, Baucke FGK, Brett CMA, Cames MF,
Covington AK, Milton MJT, Mussini T, Naumann R, Pratt KW,
Spitzer P, Wilson GS (2002) Pure Appl Chem 74:21692200
26. Mrissy M, Pratt KW, Spitzer P (2009) Metrologia 46:199213
27. Hamer WJ, Acree SF (1939) J Res Nat Bur Stand (US) 23:647662
28. Bates RG, Acree SF (1943) J Res Nat Bur Stand (US) 30:129155
29. Hamer WJ, Acree SF (1944) J Res Nat Bur Stand (US) 32:215221
30. Bates RG, Guggenheim EA (1960) Pure Appl Chem 1:163168
31. Spitzer P (2001) Accred Qual Assur 6:5560
32. Bates RG, Robinson RA (1980) Solution Chemistry 9:455456
33. OIML (1980) pH Scale for Aqueous Solutions, International
Recommendation No. 54
34. Draft International Standard ISO/DIS 26149 Water quality (2000)
Determination of pH, ISO, Geneva
35. DIN 19266 (2000) pH-measurement - Reference buffer solutions for
the calibration of pH measuring equipment. Beuth Verlag, Berlin

J Solid State Electrochem (2011) 15:6976

36.
37.
38.
39.
40.
41.
42.
43.
44.

45.
46.

47.

48.

49.

Spitzer P, Werner B (2002) Anal Bioanal Chem 374:787795

Baucke FGK (1977) Chem-Ing-Tech 49:739740
Baucke FGK (1979) Electrochim Acta 24:9597
Baucke FGK (1994) J Electroanal Chem 68:6775
Breitenbach M, Echkardt E, Einert P, Moritz M (1980) Mitteilungsblatt der Chemischen Gesellschaft der DDR 27(10):209214
CMCs in pH, BIPM Data base, Available at http://kcdb.bipm.org/
AppendixD/default.asp (date of access : 30 May 2010)
Baucke FGK (1993) Anal Chem 65:32443251
Baucke FGK (1994) Anal Chem 66:45194524
DIN 19268:2007-05 (2007) pH-measurement of aqueous solutions
with pH measuring chains with pH glass electrodes and evaluation of
measurement uncertainty
ISO 10523:2008 (2008) Water qualitydetermination of pH, ISO,
Geneva (date of access: 20 April 2010)
De Bievre P, Dybkaer R, Fajgelj A, Hibbert DB (2008) Draft
IUPAC recommendation, Available at http://old.iupac.org/proj
ects/2001/2001-010-3-500.html (date of access: 30 May 2010)
IUPAC Project: comparable pH measurements by metrological
traceability, 2005, Available at http://www.iupac.org/web/ins/
2004-005-2-500 (date of access: 20 April 2010)
Project 843 of EURAMET (European Association of National
Metrology Institutes), Available at: http://www.euramet.org/index.
php?id=tc-projects (date of access: 20 April 2010)
Guide to the expression of uncertainty in measurement (2008)
ISO/IEC Guide 98-3:2008 [JCGM 100:2008] Available at: http://
www.bipm.org/en/publications/guides/gum.html