Anal Bioanal Chem (2002) 374 : 141–144
DOI 10.1007/s00216-002-1462-0
O R I G I N A L PA P E R
J. F. van Staden · Mulalo G. Mashamba · Raluca I. Stefan
An on-line potentiometric sequential injection titration process analyser
for the determination of acetic acid
Received: 14 February 2002 / Revised: 20 June 2002 / Accepted: 25 June 2002 / Published online: 10 August 2002
© Springer-Verlag 2002
Abstract An on-line potentiometric sequential injection
titration process analyser for the determination of acetic
acid is proposed. A solution of 0.1 mol L–1 sodium chlo-
ride is used as carrier. Titration is achieved by aspirating
acetic acid samples between two strong base-zone vol-
umes into a holding coil and by channelling the stack of
well-defined zones with flow reversal through a reaction
coil to a potentiometric sensor where the peak widths
were measured. A linear relationship between peak width
and logarithm of the acid concentration was obtained in
the range 1–9 g/100 mL. Vinegar samples were analysed
without any sample pre-treatment. The method has a rela-
tive standard deviation of 0.4% with a sample frequency
of 28 samples per hour. The results revealed good agree-
ment between the proposed sequential injection and an
automated batch titration method.
Keywords Potentiometric sequential injection titration ·
Acetic acid · Sodium hydroxide · Process analyser
Introduction
Acetic acid is one of the most important intermediate and
frequently used aliphatic carboxylic acids. It is the princi-
pal acid constituent of vinegars (5% w/v) and volatile
acidity in wines [1]. Vinegar is mainly used by consumers
for acidification of salads and vegetables and for season-
ing meat and fish. The food industry uses vinegar to pre-
serve and season food at the same time [2]. Vinegar pro-
duction is carried out mainly by oxidation of alcohol-con-
taining liquids utilizing an enzyme that is present in the
vinegar bacteria (Acetobacter sp.) [3]. It can also be pre-
pared by diluting synthetically produced acetic acid.
Acetic acid is also a corrosive acid [4] with numerous in-
dustrial applications where it is important as a solvent and
J.F. van Staden (✉) · M.G. Mashamba · R.I. Stefan
Department of Chemistry, University of Pretoria,
Pretoria 0002, South Africa
e-mail: Koos.vanStaden@chem.up.ac.za
as a reagent in the production of rubber, plastics, acetate
fibres, pharmaceuticals, dyes and textiles. The determina-
tion of acetic acid in vinegar is therefore not only of criti-
cal importance in the food industry, but also in industrial
applications.
Several methods have been proposed for the determi-
nation of acetic acid in vinegar, for example gas phase
molecular absorption spectrometry [5] and a piezoelectric
crystal sensor [6] on which humidity, pressure and tem-
perature needs to be controlled. All these methods are based
on compound volatility.
Simultaneous automatic determination of acidity [7] and
a multi-commutated unsegmented flow system [8] with
potentiometric detection are among the methods reported
involving continuous flow systems. Batch titrators are
usually characterized by a low sample throughput and
high implementation cost. The concept of flow injection
(FI) titration was introduced in 1977 [9] and has since
been described in the determination of acetic acid [2, 3,
10, 11, 12]. However, the FI technique does not offer the
robustness needed in process control and has a higher
reagent consumption that makes the method expensive.
The introduction of sequential injection (SI) titration in
1997 [13] overcame this drawback by replacing the con-
tinuous flowing titrant stream of the flow injection titra-
tion system with two titrant zones on either side of the
sample zone in a distilled water carrier stream. Maskula et
al. [14] described a spectrophotometric sequential injec-
tion titration system of a weak acid in which peak area in-
stead of peak width was measured and in which an indi-
cator was used as a colour reagent. Although the sample
throughput in sequential injection systems is lower than in
FI titration, the advantages of sequential injection titration
are that the system is fully computerized, very low amounts
of titres of reagents are consumed and it is inexpensive.
This paper presents an SI titration system of a weak
acid using a pH electrode, replacing the colour indicator,
and its application to the determination of acetic acid in
vinegar samples. The concept is based on the sandwiching
of the weak acid sample zone between two strong base
zones. The main advantages of the detector are that the re-
142

sults are reliably obtained regardless of the sample colour The results obtained with the proposed system were compared
of the materials and the more cost-effective use of the in- with results obtained from an automated batch titration system. An
automatic burette (Metrohm Dosimat 665) was used for automated
strumentation. titration.

Experimental Reagents and solutions

Apparatus All solutions were prepared with chemicals of analytical grade.

The sequential injection titration system is illustrated in Fig. 1. It
was constructed from Gilson Minipuls 3 peristaltic Pump (Villiers-
le Bel, France), a 10-port electrically actuated selection valve
(Model ECSD10P, Valco Instrumentation, Houston, Texas, USA)
and a pH electrode (Hamilton, Switzerland) connected to an Orion
Research Process Ionalyzer (Model 901). Data acquisition and de-
vice control were achieved by using a PC 30-B interface board
(Eagle Electric, Cape Town, South Africa) and an assembled dis-
tribution board (MINTEK, Randburg, South Africa). The flowTEK
[14] software package (obtainable from MINTEK) for computer-
aided flow analysis was used throughout for device control and
data acquisition. Tygon tubing was used for the construction of the
holding (HC) and reaction (RC) coils. The length and diameter of
the tubing used are indicated in Fig. 1.
De-ionised water from a Modulab system (Continental Water Sys-
tems, San Antonio, TX, USA) was used to prepare all solutions
and dilutions. A 0.1 mol L–1 sodium chloride used as a carrier for
constant ionic strength and was prepared by dissolving 11.688 g of
dried sodium chloride in 2 L of distilled water. Standard sodium
hydroxide working solutions were prepared by appropriate dilution
of 0.92 mol L–1 sodium hydroxide stock solution (Titrisol standard
solution, Merck) standardized against potassium hydrogen phthal-
ate (Merck). Acetic acid standards were prepared by appropriate
dilution of 17.643 g/100 mL glacial acetic acid (Merck) stock solu-
tion in the range 1–9 g/100 mL standardised against the 0.92 mol L–1
sodium hydroxide solution. Six vinegar samples were directly
analysed without any sample pre-treatment.
Procedure

Figure 1 shows a schematic diagram of the system, with the basic

Fig. 1 A Schematic representation of the sequential injection titra-
tion system. B Device sequence for one cycle of the sequential in-
jection titration system
device sequence given in Table 1. The titration first involved the
aspiration of a base zone for 12.0 s, then an acid sample zone for
3.0 s and finally the second base zone for 6.0 s into the holding coil
as depicted in Fig. 1B. The stack of zones in the holding coil was
swept by flow reversal via the reaction coil through the pH elec-
trode as sensor. The time needed to complete one cycle was 130.0 s
and the pump returned to its starting position ready for the next cy-
cle. The sample frequency is 28 samples h–1. The peak width was
measured at a relative peak height of 1250. In the more advanced
system, port 10 of the selection valve was used as detector line and
port 9 of the selection valve as the line for the second sodium hy-
droxide stream. Valve positions 2–8 were used either for the selec-
tion of different acetic acid standard solutions or for different sam-
pling points making the system ideally suitable for automation,
quality control and on-site monitoring. The computer was then
programmed for the following device sequence: (i) Valve position
1 for the first sodium hydroxide solution, valve position 2 for the
standard or sample line and valve position 9 for the second sodium
hydroxide solution aspirated into the holding coil, and then valve
position 10 for flow reversal of the stack of zones through the re-
action coil to the sensor for measurement. (ii) The same device se-
quence was then followed in which valve position 2 in a series of
runs was replaced for the different runs by valve positions 3 to 8
for the different sampling points or standard solutions as required.
Up to three standard acetic acid solutions were aspirated in be-
tween the different samples to assure quality control of the system.

Table 1 Device sequence for

one cycle of the sequential Time (s) Pump Valve Description
titration system 0 Off NaOH Pump off, select NaOH solution (valve position 1)
1 Reverse Draw up NaOH solution
13 Off Pump stop
14 Sample Select acetic acid stream (valve position 2)
15 Reverse Draw up acetic acid sample
21 Off Pump stop
22 NaOH Select NaOH stream (valve position 3)
23 Reverse Draw up NaOH solution
29 Off Pump stop
30 Detector Pump select detector line (valve position 4)
31 Forward Pump stack zones to the detector
130 Off Home Pump off, return valve to position 1
 143

Results and discussion

The different titrant and titrand volumes were aspirated

into a holding coil where after the well-defined stack of
zones were channelled through a reaction coil to an electro-
chemical cell for detection. A carrier stream of 0.1 mol L–1
sodium chloride solution was used. The two base zones
and acid zone penetrated and gradually neutralized each
other on the boundaries when the flow of the carrier
stream was reversed and the zones were propelled to the
detector. An element of fluid therefore existed at both the
front and tail of the dispersed acid sample zone within
which the acid was exactly neutralized by the base. The Fig. 3 Influence of base volume on peak width and % RSD
theoretical concepts behind this have been discussed else- ([NaOH]=2%; [acetic acid]=5%)
where [13, 15, 16].

Method optimisation

Flow rate, zone volumes and base concentration are param-

eters that have an influence on the response and precision
of the system and they have been optimised sequentially.
A concentration of 2 g/100 mL sodium hydroxide and
5 g/100 mL acetic acid were used in the optimisation of
the flow rate. Flow rates ranging between 1.9 and 2.6 mL
min–1 were studied, and the influence on peak width and
% relative standard deviation (RSD) is given in Fig. 2.
The peak width of the resultant peak decreased with in- Fig. 4 Influence of sample volume on peak width and % RSD
creasing flow rates. The best precision was found at a ([NaOH]=2%; [acetic acid]=5%)
flow rate of 2.27 mL min–1 that was used in the optimisa-
tion of the other parameters. The influence of base vol- ume of 227 µL was selected as optimum indicating that a
ume was studied in the range of 200–800 µL (Fig. 3). The larger volume is needed for the weaker acid as compared
volume of base zone 1 was always twice that of zone 2 to to the stronger acid [17]. The linear titration range was
have two peaks of equal heights for base 1 and base 2, re- affected to a large extent when the concentration of the
spectively, after dispersion [13]. Precision was not af- sodium hydroxide changed. A concentration of 0.4 g/
fected at lower base volumes, but deteriorated at higher 100 mL sodium hydroxide solution was used for the titra-
volumes. A volume of 454 µL was chosen as optimum be- tion of 0.7–5 g/100 mL of acetic acid, 1.2 g/100 mL so-
cause of the lower % RSD (Fig. 3). Different sample vol- dium hydroxide solution for 1–9 g/100 mL acetic acid and
umes were investigated in the range between 150–350 µL 2 g/100 mL sodium hydroxide solution for 5–13 g/100 mL
(Fig. 4). The peak width increased with increasing sample acetic acid. A concentration of 1.2 g/100 mL sodium hy-
volumes. Precision was not good at lower volume due to droxide solution was used in the evaluation of the method.
low sensitivity, but was better at higher volumes. A vol-

Method evaluation

The optimised sequential injection system was evaluated

Fig. 2 Influence of flow rate on peak width and % RSD ([NaOH]=
2%; [acetic acid]=5%)
with regard to linearity, accuracy, precision and sample
frequency. The samples were analysed without any pre-
treatment.
The sequential injection titration peaks for different
acetic acid concentrations are shown in Fig. 5. The dotted
line indicates the height at which the peak width was mea-
sured. The calibration curve for the sequential injection
titration system is linear in the range 1–9 g/100 mL with
r=0.9991 and the relationship between peak width (W)
and concentration of the acetic acid (C) given by: W=
15.142logC+12.308. The accuracy of the proposed sys-
tem was evaluated by comparing the results obtained with
144

The comparison was done to establish whether the SI

Fig. 5 SI titration peaks of different acetic acid solutions titrated
with 1.2 g/100 mL sodium hydroxide solutions, the peak width
measured at 1250 (A 1 g/100 mL, B 3 g/100 mL, C 5 g/100 mL,
D 7 g/100 mL and E 9 g/100 mL)
titration system could be accepted as giving reliable re-
sults for the titration of acetic acid. The null hypothesis
that assumes that the quantities from two different meth-
ods are the same was used to check if the results from
these methods would agree. The t-test [18, 19] with mul-
tiple samples (paired by differences) was applied to ex-
amine whether the two methods differed significantly at
the 95% confidence level. The null hypothesis is Ho: ζ=0,
against the alternative Ht: ζ≠0, where ζ is the population
paired difference. The test is two tailed, as we are inter-
ested in both xd <0 and xd >0. The t-data paired test √analy-
sis was performed using the formula: tcalc = |xD | × sxn . At
the 95% confidence level, t0,005,5=2.571 and tcalc was found
to be 1.712. Since this value lies below the critical value,
±2.571, the proposed method can be accepted because the
results are comparable at the 95% confidence level.

the sequential injection titration system for the vinegar

samples with those obtained from an automated batch Conclusion
titration system. The results given in Table 2 reveal good
agreement between the sequential injection and auto- The potentiometric SI titration system described, provides
mated batch titration method. The precision of the pro- a simple, reliable and inexpensive method for the deter-
posed SI titration system was evaluated by ten repetitive mination of acetic acid in vinegar. The method has a sam-
analysis of a number of vinegar samples and standard so- ple frequency of 28 samples h–1 with a linear range of
lutions. A relative standard deviation of less than 0.4% 1–9 g/100 mL. It is easy to use, fully computerized and gives
was obtained as shown in Tables 2 and 3. It took 130 s results that are comparable with a standard automated
to complete one cycle, result in a sampling frequency of batch titration.
28 samples h–1.
References
Statistical comparison
1. Ebner H, Folman H (1983) In: Rehm HJ, Reed G (eds) Bio-
technology, XV. Weinheim
The proposed system was further validated by a statistical 2. Tubino M, Barros FG (1991) J Assoc Off Anal Chem 74:346
comparison between the proposed SI titration system and 3. Becker T, Kittsteiner-Eberle R, Luck T, Schmidt HL (1993)
a standard automated batch titration method (Table 2). J Biotechnology 31:267
4. DECHEMA corrosion handbook (1990) Behrens D (ed) Corro-
sive agents and their interaction with materials. DECHEMA
Deutsche Gesellschaft Fur Chemisches Apparatewesen, Chemi-
Table 2 Comparison of results for vinegar samples determined by sche Technik und Biotechnologie D-6000 Frankfurt am Main. 6
the SI titration system and a standard automated batch titration 5. Sanz-Vincente I, Cabredo S, Galban J (2000) Mikrochim Acta
134:199
Sample SI % RSDa Automated % RSDa 6. Levya JAM, Decisneros LHH, Debarreda DGG (1993) Talanta
titration titration 40:1725
(g/100 mL) (g/100 mL) 7. Lapa RAS, Lima JLFC, Perez-Olmos R, Ruiz MP (1995) Food
Control 6:155
White vinegar 4.616 0.27 4.66 0.321 8. Almeida CMNV, Lapa RAS, Lima JLFC, Zagatto EAG,
White vinegar 4.934 0.3 4.876 0.353 Araujo MCU (2000) Anal Chim Acta 407:213
White vinegar 4.898 0.398 4.858 0.421 9. Ruzicka J, Hansen EH, Mosbaek H (1977) Anal Chim Acta 92:
235
Brown vinegar 5.474 0.154 5.452 0.336 10. Capelato MD, dos Santos AM, Fatibello-Filho O (1996) Anal
White vinegar 5.038 0.236 5.014 0.313 Lett 29:711
White vinegar 4.948 0.211 4.888 0.436 11. Honorato RS, Araujo MCU, Veras G, Zagatto EAG, Lapa
RAS, Lima JLFC (1999) Anal Lett 15:665
an=10.
12. Martelli PB, Reis BF, Korn M, Lima JLFC (1999) Anal Chim
Acta 387:165
Table 3 Precision of the SI 13. van Staden JF, du Plessis H(1997) Anal Commun 34:147
titration system Standards % RSD 14. Maskula S, Nyman J, Ivaska A (2000) Talanta 52:91
(g/100 mL) (n=10) 15. Marshall GD, van Staden JF (1992) Process Contol Qual 3:251
16. van Staden JF, du Plessis H(2000) Talanta 52:83
1.0 0.39 17. van Staden JF, Mashamba MG, Stefan RI (2001) 55:(in press)
3.0 0.161 18. Christian GD (1994) Analytical chemistry, 5th edn. Wiley and
5.0 0.324 Sons
7.0 0.36 19. McCormack D, Roach A(1987) Measurement, statistics and
9.0 0.257 computation. Analytical chemistry by open learning. Wiley,
pp 239–240