SEflSORs
ACT@Ons
ELSEVIER
Sensors and Actuators B 38-39 (1997) 330-333
Studies on quenching of fluorescence of reagents in aqueous solution
leading to an optical chloride-ion sensor
A. Martin, R. Narayanaswamy *
Department
of Instrumentation
and Analytical Science, UMJST, PO Box 88, Manchester M60 IQD, UK
Abstract
The sensitivity of quinoline- and acridine-type indicators towards the detection of low levels of chloride ions in aqueous and amine-doped
media, by fluorescence quenching, is reported in this paper. Solution studies are based on the fluorescence of quinine sulphate, acridine and
3- (6-methoxyquinolino) propanesulphonate (SPQ) and the fluorescence responseof these indicators has been examined at different pH values
and at various chloride ion concentrations in aqueous media.
Keywords:
Chloride ions; Fluorescence quenching; SPQ
1. Introduction
The importance of chloride-ion detection involves a wide
area including pure, environmental, bio- and industrial chemistry. Industrial monitoring of chloride ions in feedwater is
necessary since chloride ions are a major cause of corrosion
in the metal components of steam-generating systems. Cer-
tain industrial specifications require the limit of detection for
chloride ions in feedwater to be 1.7-57 FM and the feedwater
typically contains amines, dissolved oxygen, carbon dioxide
and trace levels of copper and iron with a pfl range of 5-10.
Many industries also requires the detection systemsto be inline for real-time process-control applications.
Traditional methods of chloride-ion detectioninvolve titra-
to the halide concentration by Ihe Stern-Volmer equation
[9]. This reaction has been adaptedfor fibre-optic chemical
sensing of chloride ions for analytical and clinical purposes
[ lo]. A mechanism involving the fluorescencequenching of
N-( 6-methoxyquinolyl) acetoethyl ester (MQAE) by chloride ions has beenutilized in the development of a fibre-optic
probe for measuring chloride in aqueoussolution [ 111.
In this paper we describe a comprehensivesolution study
based on the collisional fluorescence quenching of quinine
sulphate, acridine and 3 (6methoxyquinolino) propanesulfonate (SPQ) by low levels of chloride ions in aqueous
media. The fluorescencestudies in aqueoussolution are preliminary and indispensable stepsfor a later sensordevelopment basedon this principle.
tions using silver nitrate, silver fluoresceinate and silver chro-
mate [ 1,2]. Titrations can be time consuming with the
end-point often difficult to detect. Spectrophotometricmethods involving reagents such as mercury thiocyanateiron( III),
mercury chloranilate and mercury diphenylcarbazide can be
hazardous due to their toxicity [ 3-51.
Fluorescenceintensity sensing is an establishedanalytical
method due to its high sensitivity and selectivity for a wide
variety of analytes. Fluorescence quenching is a common
technique usedin the detection of gasesand metal ions [ 6,7].
Collisional quenching by anions, such as alkyl halides, of
quinine- and acridine-type fluorophores has been previously
reported [ 81. The fluorescence quenching has been related
* Corresponding
4911.
author. Phone: f44
161200 4891. Fax: +44 161200
0925-4005/97/$17.00
0 1997 Elsevier Science S.A. All rights reserved
PIISO925-4005(97)00044-O
2. Experimental
Quinine sulphate (BDH), acridine (BDH) and SPQ
(Sigma) were used as purchased.All quinine sulphate and
acridine solutions were prepared.in 0.05 M sulphuric acid,
while SPQ solutions were prepared in deionized water
(Elgastat). The concentration of all the Eluorophorestock
solutions was 10 PM. The stock amine (2-amino-2-methyll-propanol, AMP, (97% Aldrich) ) solution (200 mg 1-I)
was preparedin deionized water j.Yornwhich a working solution of AMP (20 mg 1-l) was made. Potassium chloride
(Aldrich) dissolved in either AMP or deionized water
was used to prepare standards of various chloride-ion
concentrations.
A. Martin, R. Narayanaswamy/Sensors
andActuators
The solution studies were carried out on the Perk&Elmer
LS-5 Luminescence Spectrometer using 90” excitation and
emission alignment. In the solution studies a fixed volume of
3.1 ml was maintained and all measurementswere performed
at room temperature. pH measurementswere made using a
hand-held meter (Mettler Toledo-Checkmate 90).
331
B 38-39 (1997) 330-333
The fluorescenceintensities of fluorophore solutions have
been experimentally observed to stabilize and then decrease
over a period of time. For fluorescence quenching the fluorophore solution should not be degrading, hence it is important to use the fluorophore solution only when the
fluorescenceintensity is stable. The aim of this study was to
determine experimentally the time it takes for the fluorescence of the solution to stabilize and therefore the optimum
time for use.All solutions were stored in the dark to minimize
any photobleaching effects due to ambient light. The fluorescence intensities of quinine sulphate, acridine and SPQ prepared in their appropriate solvents were monitored at 24 h
intervals over a 96 h period. After 24 h the intensities of all
the fluorophores were noted to increase.For acridine, aperiod
of 24 h appeared to be the optimum age. The excitation and
emission intensities of SPQ and quinine sulphate remained
relatively constant after 24 h. While SPQ appearedto be the
most stable fluorophore, acridine was the least stable. It was
therefore decided to use only fluorophore solutions that have
been stored for a period of 24 h for further studies.
To determine the most sensitive fluorophore towards low
levels of chloride ions, a fluorescence quenching study was
carried out. By using the Stern-Volmer equation
3. Results and discussion
Both quinine sulphate and acridine exhibit fluorescencein
ethanol, but to be selectively quenchedby haiide ions (chloride) they must be in their protonated form. Both indicators
were protonated in 0.05 M sulphuric acid as describedpreviously [8,12]. Quinine sulphate will not fluoresce in
hydrochloric acid due to its fluorescence quenching effect
[ 131. SPQ is a neutral fluorophore and is therefore not pH
dependent. It can fluoresce in acid, alkali, alcohol or water.
Thus SPQ solutions were prepared in deionized water.
To determine the effect of amine-doped feedwater on the
fluorescenceofthe indicators, excitation andemissionspectra
were recorded in its presence. These spectra compared
favourably with previous work [9,14]. Table 1 shows the
excitation and emission wavelengths for the indicators.
The AMP working solution (20 ppm) is alkaline, whereas
deionized water is almost neutral (pH 6.30). The pH of 0.05
M sulphuric acid is 1.3. Table 2 shows the change in pH of
the indicators in their appropriate solvents when added to
AMP and deionized water. The sulphuric acid in which quinine sulphate and acridine are prepared protonates the weak
AMP, The pH of AMP was not affectedwhen SPQwas added.
No spectral shift or change in peak shapewas observedin the
spectra, indicating that the presence of AMP in the solvent
had no effect on the fluorophores. A shift in wavelength of
the emission spectrum observed in different solvents is a
common phenomenon when solvent polarity is changed.
Water is more polar than alcohol, but in this case AMP is
present at very low concentrations (20 ppm) and hence the
solvent properties are more characteristic of water.
F,/F=
Ksv[Q]
+ 1
where F,, is the fluorescence intensity in the absenceof the
quencher, F is the fluorescence intensity in the presenceof
the quencher, KS, is the Stern-Volmer quenching constant
and Q is the concentration of the quencher, calibration graphs
for quinine sulphate, acridine and SPQ were obtained
(Figs. l-3). From thesecalibration graphsthe detectionlimit
of chloride ions, defined as the concentration equivalent to
F,,/ (F. - 3sd) where sd is the standarddeviation of the blank
( Fo) , could be calculated. The detection limits, although theoretical values [ 15,161, are valuable as a comparison of the
different detection processesemployed. Therefore, the detection limits for the fluorescencequenching processof quinine
sulphate, acridine and SPQ with the experimental values of
the lowest measurableconcentration of chloride ions and the
Table 1
Excitation and emission wavelengths of quinine sulphate, acridine and SPQ
Indicator
Quinine sulphate
(0.05 M H,SO,)
Aqueous medium
Excitation wavelength (nm)
Emission wavelength (nm)
A&dine
(0.05 M H,SO,)
SPQ
(water)
water
AMP
water
AMP
water
AMP
360
465
362
465
367
49.5
367
492
330
463
330
465
Table 2
pH measurements for quinine sulphate, acridine and SPQ
Indicator
Solvent
Aqueous medium
PH
AMP
9.79
QS
AC
SPQ
QS
H&b
AC
SPQ
Hz%
H,SO,
H,SO,
H,O
H,O
H804
AMP
HzW
H2O
J320
WO4
AMP
H,SO,
r-l,0
Hz0
AMP
Hz0
Hz0
6.30
1.33
1.31
5.37
2.47
2.32
2.46
2.38
9.30
5.22
332
A. Martin, R. Narayanaswamy/Sensors
!b
2'5
3'0
3'5
4'0
45
and Actuators 3 38-39 (1997) 330-333
0.5-l
60
0
I
5
‘
10
I
15
cmc chloride ion (ppm)
Fig. 1. Collisional quenching of quinine suiphate by chloride ions. (Solid
and dashedlines are the regression plots.)
1.170-
+
-
1.145-
-
N&r
y,+&?r
A
A$p
......' .Qfp
1.120-
0
5
10
15
20
25
30
35
20
*
25
2
30
*
35
I
40
45
Mnc chloride ion?3(ppm)
40
45
50
ME chloride ions (ppm)
Fig. 2. Collisional quenching of acridine by chloride ions. (Solid and dashed
lines are the regression plots.)
I&v values are given in Table 3. The lowest measurableconcentration values are higher than the detection limits. The
lowest measurable concentrations of chloride ions are real
values and hence more useful in comparing the different
detection processesemployed in the working system. The
KS, values indicate the order of sensitivity of the different
fluorophores for chloride ions. The larger the KS, value, the
more sensitive the fluorophore is to chloride ions. The order
of sensitivity was found to be SPQ>quinine sulphate
B acridine.
The fluorescence quenching of quinoline and acridine
derivatives, such as SPQ, by halide ions is not a novel phenomenon, and hasbeenpreviously usedin solution andsensor
applications. Wolfbeis and coworkers [lo] applied this
quenching system to develop an optical sensorfor continuous
monitoring of halide ions with a KS, value of 118 M- ’ for
chloride ions and a limit of detection of 10 r&I. This KS,
value was around lO-20% lower than thoseof the non-immo-
Fig. 3. Collisional quenching of SPQ by chloride ions. (Solid and dashed
lines are the regression plots.)
bilized indicator in solution. This reduction was explained by
immobilization limiting the mobility of the fluorophore and
therefore lowering the probability of the quenching process
occurring. Krapf et al. [ 171 used SPQ to detect intercellular
chloride ion via fluorescencequenching, with a KS, value of
118M - ‘. Kar and Arnold have determined the KS, for chloride-ion quenching of MQAE as 199M - ’ [ 111. In our work
we have found that the sensitivity of SPQ quenching to chloride ions (KS,) was greatly enhanced (717 M-l) without
the aid of preconcentration techniques,which could be agreat
advantage for sensing applications.
It is well known that bromide and iodide quench the fluorescenceof the dyes investigated in this work much more
strongly than the chloride ion [ lo]. However, interference
studies were not carried out in this investigation due to the
fact that the feedwater contains only chloride ions.
4. Conclusions
The presenceof AMP at low concentrations did not affect
the fluorescencequenching of quinine sulphate, acridine and
SPQ. While SPQis a neutral fluorophore, the fluorescenceof
quinine sulphateand acridine were quenchedby ch1orid.eions
in their acidified forms. The dilute sulphuric acid in which
quinine sulphate and acridine were prepared protonated the
AMP solution; hencequenching by chloride ions occurred.
Quinine sulphate and SPQ were found to be photostable,
whereas the fluorescenceintensity of acridine was found to
decreaseafter 24 h. Subsequentlyall solutions were stored in
the dark for 24 h prior to use,therefore letting thelhiorophores
diffuse and equilibrate in solution.
Table 3
Detection limits and Stem-Volmer quenching constants for quinine sulphate, acridine and SPQ
Indicator
Aqueous medium
Detection limit (mM)
Lowest measureable concentration (mM)
f&v Of-‘)
Quinine sulphate
(0.05 M H2S04)
water
0.5
0.7
213
Acridine
(0.05 M H,SO,)
AMP
0.6
0.8
213
water
0.9
1
112
SPQ
(water)
AMP
1
1.1
111
water
0.1
0.3
717
AMP
0.1
0.3
717
A. Martin, R. Narayanaswamy/Sensors
SPQ was found to be the most sensitive fluorophore with
the lowest calculated detection limit and the lowest measurable chloride-ion concentration. Immobilization of a lumiphore onto solid supports can enhance the rigidity of the
molecule and improve its luminescence characteristics, such
as intensity and quantum yield [ 181, thus increasing the sensitivity of the lumiphore. This can lead to lower detection
limits of the analyte to be determined. Solid surface analysis
is most commonly applied to room-temperature phosphorescence but can also be applied for fluorescence. Therefore
studies involving the immobilization of SPQ with a view to
lowering the detection limit of chloride ions are being currently performed. A detection of chloride ion in the range
1.7-57 p,M in feedwater is required for the intended industrial
application. Investigations on the physical immobilization of
SPQ onto anion-exchange resins and chemical immobilization by covalent bonding to amine gels and entrapment into
sol-gels are currently in progress with a view to developing
an optical chloride-ion sensorwith the abovedetection limits.
Acknowledgements
The authors acknowledge Rolls Royce and AssociatesLtd.
for their financial support of this project.
References
[l] AI. Vogel, A Text-book of Quantitative
inorganic
Analysis,
Longmans, London, 3rd edn., 1961.
[2] F.D. Snell, PhotometricandFIuorometric
Methods ofAnalysis-Nonmetals, John Wiley, Chichester, 1981.
[3] D.F. Boltz and J.A. Howell, Colourimetric
Determination
of Non
Metals, John Wiley, Chichester, 1978.
[4] F.3. Krug, L.C.R. Pessenda,E.A.G. Zagatto, A.O. Jacintho and B.F.
Reis, Spectrometric flow injection determination ofchlorldeinethanol,
Anal. Chim. Acta, 130 (1981) 409413.
1.51R.E. Humphrey and W.L. Hinze, Mercuric iodate as a analytical
reagent determination of chloride by spectrophotometric
measurement of mercuric chloride with phenolphthalein complexone
or xylenol orange, Anal. Chem, 45 (1973) 1747-1749.
[6] M.G. Baron, R. Narayanaswamy and SC. Thorpe, A kineto-optical
method for the determination of chlorine gas, Sensors and Actuators
B, 29 (1995) 3.58-362.
[7] M.H. Noire and B. Dureault, A ferrous ion optical sensor based on
fluoresence quenching, Sensors and Actuators 3,29 (1995) 386-391.
and Actuators 3 38-39 (1997) 330-333
333
[S] OS. Wolfbeis and E. Urbana, Fluorescence quenching method for
determination of two or more components in solution, Anal. Chem.,
55 (1983) 1904-1906.
[9] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, Plenum
Press,London, 1983.
[lo] E. Urbano, H. Offenbacher and O.S. Wolfbeis, Optical sensor for
continuous determlnation of halides, Anal. Chem., 56 (1984) 427429.
[ll]
S. Kar and M.A. Arnold, Fiber-optic chlorine probe based on
fluorescence decay of N-(B-methoxyquinolyl)-acetoetbyl
ester,
Talanta, 42 (1995) 663-670.
[ 121 R.A. Velapoldi, Considerations on organic compounds in solution and
inorganic ions in glasses as fluorescent standard reference materials,
.J. Rex Nat. Bar. Stand. -A. Phys. Chem., 76A (1972) 641-654.
[ 131 G.G. Stokes, Onacertain reaction ofquinine, J. Chem. Soc.,ZZ (1869)
174-185.
[ 141 LB. Berlman, Handbook of Fluorescence Spectra of Aromatic
Molecules, Academic Press,London, 2nd edn., 1971.
[ 151 G.L, Long and J.D. Wineforder, Limit of detection - a closer look at
the IUPAC definition, Anal. Chem., 55 (1983) 713A-724A.
[ 163 J.C. Miller and J.N. Miller, Statistics for Analytical Chemistry, Ellis
Horwood, London, 3rd edn., 1993.
[ 171 R. Krapf, C.A. Berry and AS. Verkman, Estimation of intercellular
chloride activity in isolated perfused rabbit proximal convoluted
tubules using a fluorescent indicator, Biophys. J., 53 (1988) 955-962.
[IS] A. Sanz-Medel, Solid surface photoluminescence aud flow analysis: a
happy marriage, Anal. Chim. Acta, 283 (1993) 367-378.
Biographies
A. Martin obtained her I3.Sc. in chemistry with information
technology and instrumentation in 1993 at Glasgow Caledonian University. She is currently studying towards a Ph.D.
at the Department of Instrumentation and Analytical Science
(DIAS) at UMIST.
R. Narayanaswamy is currently a senior lecturer in DIAS
at UMIST. He obtained his Ph.D. in 1972 (Imperial College,
London, UK) in analytical chemistry and his D.Sc in 1995
(University of London) in analytical science. He was a lecturer in chemistry at the University of Sri Lanka, Peradeniya,
Sri Lanka (1967-1978) and subsequently a postdoctoral
research fellow at the University of Southampton, UK
(1978-1981) and at the University of Warwick, UK ( 1982).
He joined UMIST, UK, in 1983 as a senior postdoctoral
research associate and became manager of the Optical Sensors ResearchUnit (1984-1987) and lecturer in instrumentation and analytical science (October 1984). He leads the
research group in DIAS which deals with fundamental and
applied studies in molecular spectroscopy and chemically
sensitive optical-fibre sensorsand devices.