- Email: [email protected]
or carbon dioxide-sensitive optical transducer
~
Pergamon
0039-9140(95)01436-5
Talanta, Vol. 42, No. 3, pp. 483~,92, 1995 Copyright © 1995 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0039-9140/95 $9.50 + 0.00
A NOVEL OXYGEN AND/OR CARBON DIOXIDE-SENSITIVE OPTICAL TRANSDUCER MING FAT CHOI* a n d PETER HAWKINS Faculty of Applied Sciences, University of the West of England, Coldharbour Lane, Frenchay, Bristol BSI6 IQY, U.K.
(Receh~ed 2 September 1994. Accepted 12 September 1994)
Summary A novel oxygen (02) and/or carbon dioxide (CO 2)-sensitive transducer for the measurement of both gaseous 02 and CO., over the concentration ranges of O~, 0 100'% and CO,, 0 10% has been described employing a solution of 10.6 llM fluorescein (FL) and 190 FLMpotassium hydroxide in a solvent mixtures of 1:1 (v/v) N,N'-diethylaniline (DEA) and N,N-dimethylformamide. Increasing 02 concentrations cause the absorbance of the solution at a wavelength of 400 nm to increase owing to a contact charge transfer reaction existing between 02 and DEA molecules, and increasing CO~ concentrations produce a non-linear fall in absorbance at 520 nm as the colour of FL changes from its orange dianion form to the colourless neutral, lactonic form. Both processes are independent of each other and reversible. The response to changes in 02 concentrations is in good agreement with Beer Lambert's law and the response to changes in CO 2 concentrations in non-linear. A fibre optic sensing system based on this solvent-
O x y g e n (O2) can be considered to be one o f the m o s t i m p o r t a n t elements in o u r n a t u r e with an e n o r m o u s chemical a n d b i o c h e m i c a l reactions involving 02 as r e a c t a n t s or products. C a r b o n d i o x i d e (CO2) is usually a b y - p r o d u c t o f m a n y industrial processes a n d it is believed to aggravate the g r e e n h o u s e effect o f o u r e n v i r o n m e n t . M e d i c a l diagnosis a n d t r e a t m e n t o f critically ill patients in intensive care units a n d o p e r a t i n g theatres often requires m o n i t o r i n g o f CO2 and O: p a r t i a l pressures o f arterial b l o o d . Consequently, the d e t e r m i n a t i o n o f b o t h 02 a n d C O , is o f c o n s i d e r a b l e i m p o r t a n c e in e n v i r o n m e n t a l , b i o m e d i c a l analysis a n d analytical chemistry. C o n s i d e r a b l e effort has been d e v o t e d over m a n y years to the d e v e l o p m e n t o f new techniques for the m e a s u r e m e n t o f 02 a n d COz c o n c e n t r a t i o n s . A m p e r o m e t r i c a l d e t e r m i n a t i o n o f O, by a C l a r k - t y p e electrode ~ has been c o m m o n l y used in the past which suffers from the d r a w b a c k o f 02 c o n s u m p t i o n . The detection o f CO2 is usually based on an infrared d e t e c t o r 2"3 a n d a Severinghaus electrode 4 with b u l k y a n d expensive devices being used. W i t h the a d v e n t o f fibre optics technology, there is a g r o w i n g interest in the d e v e l o p m e n t o f fibre optics 02 a n d / o r CO2
sensors. The attractive features o f fibre optics sensors include: immunity to electrical interference; ease o f m i n i a t u r i z a t i o n ; c o m p a r a tively inexpensive device; r e m o t e a n d in situ sensing; a n d high i n f o r m a t i o n c a r r y i n g c a p a c i t y (multiwavelength transmission), etc. There have been m a n y p u b l i c a t i o n s on single optical fibre 02 sensors 5 8 based on the fluorescent q u e n c h i n g o f a dye by m o l e c u l a r 02 and C O , sensors 9 14 based on the p H m o d u l a t i o n a c c o m p l i s h i n g with the c o l o u r change o f a dye. Several electrochemical multi-sensors have also been r e p o r t e d for the s i m u l t a n e o u s l y detection o f both O, a n d CO,,. A l b e r y and B a r r o n ~5 a p p l i e d the electroreduction technique for the a m p e r o m e t r i c d e t e r m i n a t i o n o f 02 a n d CO2 in which two electrolytes and two a p p l i e d potentials were used. A r q u i n t et al. t6 a n d G u m b r e c h t et al. ~7 employed amperometric and potentiometric techniques for the detection o f pH, pO2 a n d pCO~. I n d i v i d u a l pH, pCO2 and pO2 sensors were fabricated and p o s i t i o n e d on a m i n i a t u r ized microchip. Recently, there have been a few c o m b i n e d fibre optic O~ and C O , sensors described in the literature. G e h r i c h et al. ~ a n d Miller et al. t9 a r r a n g e d three individual p H , pCO2 a n d pO2-sensitive optical fibres t o g e t h e r in a catheter. Similarly, oxygen-sensitive m a t e r i a l
*Author to whom correspondence should be addressed. tat
42 ~
t
483
484
MING FAT CHOl and PETERHAWKINS
and CO2-sensitive material were entrapped in separate gas-permeable membranes and attached to the distal end of a bifurcated optical fibre for sensing gaseous 02 and CO2.20 In all the above instances, individual sensitive-material was often used for each gas. To the best of our knowledge, not a single sensing medium for both 02 and CO2 have been reported to date. It has been known that some organic solvents exhibit contact charge transfer absorption (CCTA) bands when in contact with O2.2J Tsubomura and Mulliken22 explained that this is caused by the formation of CCT pairs between the organic solvents and molecular 02. The CCTA bands lie in longer wavelength regions if the solvents have lower ionization potentials, like some amine compounds, and usually appear as extensions to the long wavelength edge of the absorption spectra of the deoxygenated solvents. The CCTA intensity of the organic solvents is also related to the partial pressure of 02 in the gas mixtures above the solvents and the reaction between the solvents and 02 is reversiblefl 3 Fluorescein (FL) dye immobilized in poly(ethylene glycol) has been successfully applied for fibre optic sensing of CO2.24 With the combination of organic solvent and dye, a single transducer can possibly be employed for the determination of both 02 and CO2. In this paper, we apply this approach and report here a single 02/C02-sensitive transducer in which both gaseous 02 and CO2 can be detected independently based on the CCTA of N,N '-diethylaniline (DEA) with 02 and the colour change of FL upon exposure to CO2. EXPERIMENTAL
Chemicals and reagents DEA (>99%), N,N-dimethylformamide (DMF > 99.9%, HPLC grade), n-heptane (99%, spectrophotometric grade), methanol (>99.9%, HPLC grade) and FL dye (98%) were purchased from Aldrich Chemical Co. Ltd, U.K. Potassium hydroxide (KOH, AnalaR grade) was obtained from BDH Chemical Ltd, U.K. Solutions of 10.6 /~M FL and 190 /~M KOH in DMF, and 10.6 pM FL and 190 /~M KOH in 1:1 (v/v) DEA/DMF solvent mixture were prepared by adding 20 ~1 of a concentrated methanolic KOH solution (KOH concentration was determined by standard acid-base titration 25) and 70 /~1 of a concentrated FL in DMF solution (concentration of FL solution
was determined spectrophotometrically by employing the molar absorptivity of FL 2dianion 26) into DMF and !:1 (v/v) DEA/DMF solvent mixture, respectively. Similarly, a solution of 0.10 #M FL and 190/~M KOH in 1:1 (v/v) DEA/DMF solvent mixture was prepared for fluorescence studies. All reagents were used as received. Nitrogen gas ( S 2 ) , 0 2 gas, 10% (v/v) CO2 in N2 gas mixture and CO2 gas were supplied by Distillers MG, U.K.
The gas mixing system Different 02 (in the range of 0-100%) or CO2 concentrations (in the range of 0-10%) in a gas stream were produced by controlling the flow rates of either O2 gas or 10% CO2 in N 2 gas mixture and the diluent N 2 gas entering a mixing chamber. The gas mixture was passed through a portable 02 meter (Oxywarn 100I from Draeger Manufacturing, U.K.) or CO2 detector (LFG 10 Landfill Gas Analyser from Analytical Development Company Ltd, U.K.) where the 02 or CO2 concentrations in the gas mixture were determined before bubbling through the investigating solvent or solvent-dye solution contained in a 1 cm path-length quartz cuvette.
Instrumentation The CCTA spectra were measured on a Perkin-Elmer Lambda 15 Spectrophotometer, fluorescence excitation (EX) and emission (EM) spectra were recorded on a Perkin-Eimer LS-5 Luminescence Spectrometer. In the fibre optic detecting system of O2 and CO,,, a home-made optical arrangement was set up. A modulated beam of light with known wavelength was produced by a 100 W quartz halogen lamp (Bentham Instruments Ltd, U.K.) stabilized by a Bentham 505 current stabilized filament lamp power supply, a Bentham 218 optical chopper at a frequency of 141 Hz and a Bentham M300 monochromator controlled by a Bentham SMD3B stepping motor drive unit before being launched into a plastic optical fibre (I m long and 2 mm diameter from RS Components, U.K.). The transmitted light from the fibre irradiated the front surface of a I cm quartz cuvette containing 1 ml of solvent or solvent~zlye solution with gaseous O~ or CO, standards at a flow rate of about 50 cm3/min continuously passing through it. A second plastic optical fibre (0.5 m long, 2 mm diameter) was positioned on the back surface of the cuvette to collect the transmitted light. The other end of this fibre led directly onto the front surface of a
02/CO2-sensitive optical transducer
485
CCTA spectra for different mol percentage (%) mixtures of DEA/n-heptane mixture were also recorded (Fig. 2). There is an increase in absorbance as the mol% of DEA in DEA/nheptane mixture increases, n-Heptane has no CCTA band at the wavelengths investigated. A graph of abs at 400 nm against mol% of DEA in DEA/n-heptane mixture is plotted and a linear straight line is found which indicates that the CCTA is directly proportional to the concentration of DEA in DEA/n-heptane mixture. The change of absorbance arises from the CCTA of the DEA and 02 complex and the reaction is reversible.
100 mm 2 silicon photovoltaic detector (RS Components, U.K.) and the output signal was amplified by a Bentham 277 current preamplifier and a Bentham 223 lock-in amplifier which was synchronized to the modulated light beam by a reference signal provided by the light chopper. The amplified signal was recorded on a Linses IS chart recorder (Electroplan, U.K.) and displayed on a LED display (Bentham 217 digital unit). To avoid a slight drift caused by a small evaporation of the solvent, the solvent-dye solution was replaced with a fresh one at approximately 2 hr intervals. RESULTS AND DISCUSSION
hv
DEA +O2 ~ D E A ' • • 02 ~ ( D E A - ' • O2)*.
Development of O:/C02-sensitive transducer N,N'-Diethylaniline has a strong broad
It is possible that DEA can serve as a sensing medium for O2. It has also been observed that some pH-sensirive dyes change colour upon exposure to gaseous CO2. The presence of water is crucial for response to CO2 which is based on the following equations:
CCTA band lying in the ultraviolet/visible regions and there is an increase in absorbance as the 02 concentration increases (Fig. 1). The CCTA band disappears when the dissolved 02 is removed by purging DEA with N 2 or CO2. Note that the reference is air-saturated DEA ([02] = 21% approx.) which explains why DEA has negative absorbances when their 02 concentrations are less than the air-saturated value. A graph of absorbance (abs) at 400 nm against applied [02] is plotted where abs = absorbance of DEA saturated with a given applied [02] - absorbance of DEA saturated with N2. A linear straight line is obtained which agrees with Beer-Lambert's law having absorbance directly proportional to the applied [02]. Similarly, the
C O
2 -b
H20 ~ H2CO s
H z C O 3 + H 2 0 ~- HCO3 + H3 O+
H C O ; + H 2 0 ~ CO~- + H 3 0 + DH + H 2 0 ~ D - + H3 O+ , (colour A)
(colour B)
where D H and D - are the protonated and deprotonated forms of the dye, respectively. The
Abs 08
Slope n 0.04xi0' Intercept = 5.88xi0 ~ Correlation coefficient
06
Abs
= 0.9998
02
1.1 0
0
'
J ~0
4(I
'
~
GO
'
BO
[o21 / %
0.7
'
I00
1 : i 0 0 % 02 2 : 8 0 . 6 % O~ i n 3 : 6 3 . 3 % 02 in 4 : 4 5 . 3 % O~ i n 5 : air-saturated 6 : 1 2 . 3 % 02 in 7 : N~
0.3
N~ N~ N2 N2
k0.1
o,5
I
!
I
!
!
!
350
390
430
470
510
55O
Wavelength
(rim)
Fig. I. The effect of different 0 2 concentration on the C C T A spectrum of D E A . Inset shows the calibration plot of absorbance at 400 n m against [O2] using the data of the spectra. The optical pathlength is I cm
and the reference is air-saturated DEA.
486
M I N G FAT CHOI a n d PETER HAWKINS
Abs
06
Intercept ~ 2.15xi0 z Correlati~ coefficient = 0.9977
04
Abe
02
1.1
~ M o l e 2
%of DEAin D~n-hepmneMi~ures
0.7 3
0~9enated
100 78.6 58.0 38.0 1827
4
0.3
mole % of D E A
Deoxgenated
in D~/n-heptane mixtures ..............................................
~
5
1 2 3 4 5 6
u u
i' 2' 3' 4' 5" 6'
-0.1
4' 5" -0.5
F h ' 350
!
I
|
!
!
400
450
500
550
600
Wavelength
(nm)
Fig. 2. The CCTA spectra o f D E A / n - h e p t a n e solvent mixtures. Solvent mixtures are in mol%. Inset s h o w s the calibration plot o f absorbance at 400 nm against m o l % o f D E A in D E A / n - h e p t a n e solvent mixture. The measurement conditions are the same as in Fig. I.
colour change of the dye is sensitive to pH change of its environment. A solution of 10.6 /tM FL and 190 # M KOH in D M F was tested for its response to CO2 (Fig. 3). It is assumed that FL exists only as FL 2 ions in this highly alkaline solvent. Neutral form of FL can occur in three different tautomers, i.e. zwitterion, quinoid and lactone. However, it has been reported that colourless lactonic form is usually the dominant tautomer present in organic solvents. 27 The change of absorbance arises from the conversion of the orange dianion form of FL to the colourless, neutral lactonic form upon exposure to CO2. These changes were brought
about from the change of the pH of the solution with the dissolution of CO2 in D M F and can be envisaged in the following reactions: CO2+OH
~ HCO3
(I)
~ H20
(2)
H + + OH
2H + + FL 2- ~ H2FL.
(3)
Dissolved CO2 removed the O H - ions in DMF: as a result, the H + ion concentration is increased which shifts the equilibrium of equation (3) from left to right. A plot of abs at 520 nm against [CO2] is shown in Fig. 3 and increasing [CO2] cause a non-linear fall in
Abs /kbs
1. 28" 04
0.?2' 0"20
2
4
6
8
10
[co2] / % 0.64'
/
/
3
¢
1 : Nl 2 : 0.88 3 : 1.99 4 : 2.78 5 : 3.34 6 : 4.60 7 : 6.20 8 : 8.73 9 : 10.0
i0:
i00
% % % % % % % %
CO 2 CO 2 CO 2 CO~ CO 2 CO 2 CO~ CO~
in in in in in in in in
N2 N2 N2 N~ N3 N2 N~ Nz % CO 2 in N 2
8.36
8.88' 10
- 0 . 2 e ..................................................... •........................................................ , .......................................................... ' ....................................................... ' .........................................' 388 424 46~ 512 556 600 Wavelength
(ran)
Fig. 3. The effect o f different C O , concentration on the absorption spectrum o f a solution o f 10.6 p M FL and 190 p M K O H in D M F . Inset s h o w s the plot o f abs against [CO2] using the data o f the spectra. T h e optical pathlength is 1 cm and the reference is a solution o f air-saturated 190 ~tM KOH in DMF.
O,/C02-sensitive optical transducer (a)
487
Abs
1 .20
092"
] I
O 64
1
:
100
2 3 4 5 6 7 8 9 i0 Ii
: : : : : : : : : :
87.0 % 02 in 7 1 . 6 % 0 z in 55.9 % 02 in 4 2 . 0 % O 2 in 3 0 . 0 % O 2 in 26.3 % O~ in air-saturated 14.6 % O~ in 7.5 % o~ in N2
% % N2 N2 N2 N2 N2 N1 N2 N2
12
12 13 14 15 16 17 18 19 20 21 22
: : : : : : : : : : :
0.38 0.70 1.35 2.44 3.03 3.82 5.14 6.34 8.39 10.0 100
% % % % % % % % % % %
C02 C% CO 2 C02 CO s CO 2 C02 CO 2 COz CO 2 CO:
in in in in in in in in in in
N2 N2 N2 N2 N2 N2 N~ N2 N2 N~
0
0.08
16 17 18 19 20
-0.2e
424
380
468
21
22
51~ Wavelength
556
!
600
(nm)
[Oz] / %
(b) 0
20 '
I
40 '
I
60 I
'
,
80 ,
,
I00 ~'110'4
ill)
0.2
0
2
4
6
8
10
[CO2I / %
Fig. 4. (a) The effect of 02 and CO2 on the absorption spectrum of a solution of 10.6/zM FL and 190 FM KOH in 1: 1 (v/v) DEA/DMF solvent mixture. Optical pathlength: 1 cm; reference: air-saturated 190 FM KOH in l:l (v/v) DEA/DMF solvent mixture. (b)(i) Plot of abs against [02] at 400 nm using the data in (a). (ii) Plot of abs against [CO2] at 520 nm using the data in (a).
absorbance with a 91% change in absorbance from 0 to 10% CO2. We observed that a solution of FL in D E A changes colour on exposure to CO2 but that the change is not entirely reversible which may be probably owing to DEA being an almost completely non-aqueous solvent. As mentioned above, a solution of FL in D M F responds reversibly to gaseous CO2 as D M F contains a small amount of dissolved water (about 0.03%). Unfortunately, D M F has a CCTA spectrum which is too far into the ultraviolet region to be of practical use for future fibre optic sensor development. After further investigation, we
devised a way around this problem by using a solution of FL dye and KOH in a 1:i (v/v) D E A / D M F solvent mixture for the independent and reversible measurements of CO2 (using colour changes of the FL dye) and 02 (using the CCTA of the D E A with 02). The observed changes in absorption spectrum of the solventdye solution as a function of [02] and [CO,,] are shown in Fig. 4a which demonstrates that there are no cross-interference between the two gases. A graph of abs at 400 nm against applied [02] is plotted in Fig. 4b(i). The results agree with Beer-Lamberrs law and the absorbance is linearly proportional to the applied [02]. The
488
MING FAT CHOI and PETERHAWKINS
(a)
(b)
(c)
# 6e--
1 2 3 4 5 6 7 8
: : : : : : : :
9 : 10 : Ii : 12 : 13 14 : 15 :
48
:
"~
air-saturated 0.07 0.19 0.32 0.44 0.52 0.76
% % % % % %
CO~ CO2 CO2 CO~ C02 CO2
in in in in in in
0.99 1.79 3.22 5.06 6.44
% % % % %
CO~ CO~ C~ C02 CO~
in N 2 in N 2 N~ in N 2 in l~
N2 N~ N2 N2 N~
in
]
i0.0 % CO~ in N~ I00
6e-
12 3
N2
3 s
t CO~
"
i
~
7
~
89
-
lO
7
20
1 3 89
1~21_3~.,is I
4~O
448
400
L-
~20
Wavelength
I
--
-I
~60
1231415
8
~2e
608
(ran)
~O
Wavelength
600
i0111213 i~15
e 520
(nm)
560
Wavelength
608 (nm)
(d)
Intensity 60 _
-
. . . . . . . . . . .
° - ~
- - = _ _
50 40 -": 30
20 10 r~.___~
0o
2
T--
-.-~_._ ~ .......
4
~. . . . . . . . . . .
6
.
,
8
[C021 / % Fig. 5. Fluorescence excitation (EX) and emission (EM) spectra of 0.10/aM FL and 190 #M KOH in 1:1 (v/v) DEA/DMF solvent mixture. Optical pathlength:l cm. (a) EX spectra with EM wavelength at 538 nm. (b) EM spectra with EX wavelength at 520 nm. (c) EM spectra with EX wavelength at 490 nm. (d) Plot of intensity against applied [CO2] using the data of the spectra.
change of absorbance is again caused by the CCTA of the DEA and 02 complex. A plot of abs at 520 nm against applied [CO2] is shown in Fig. 4b(ii) and increasing [CO2] produce a nonlinear fall in absorbance. The change of absorbance arises from the conversion of the orange dianion form of FL to the colourless, neutral lactonic form upon exposure to CO2. The fluorescence EX and EM spectra of 0.10 /~M FL and 190 /tM KOH in 1:1 (v/v) D E A / D M F solvent mixture is shown in Figs. 5a-c. The fluorescence intensity is decreased when the CO2 concentration is increased. Unlike FL 2- ion which is a strongly
fluorescent molecule, the neutral lactonic form of FL is non-fluorescent. As explained above, FL 2- ions in the solution are probably converted to neutral lactonic form of FL upon exposure to CO2 (equations I-3). The plot of intensity against applied [CO2] is shown in Fig. 5d which shows that increasing [CO2] generates a non-linear fall in fluorescence intensity.
Fibre optic sensing system for 02 and CO_, The experimental results demonstrate that FL dye and KOH in D E A / D M F solvent mixture is a promising 02 and CO2-sensitive transducer to be used for the development of a sensing system
O2/CO2-sensitive optical transducer
489
(a) 10% COy
(b)
::
o o
,n
Time
(min)
(c) 0.7 slope. 4 a~xlO' Intercept - 4 4xze ~
,,
0.6
:o..,,,
....
-
,,o, ....
o-,,
.-
>~ o
"~0.4
....
0.3 0.5
~;~.-
S
"
."
0,2 <
0,4
0.3 .
t
0
0.1
,:. '~
L
20
40
60
1~ J O
80
100
[Od / %
Fig. 6. (a) The response time, reproducibility and total signal change of the fibre optic sensing system when subjected to changes between: N2---*O2---~10% CO2 in N2---.O2---~N2. 1.0 ml of a solution of 10.6 #M FL and 190 /~M KOH in 1:1 (v/v) DEA/DMF solvent mixture in a 1 cm cuvette was used with a wavelength of 400 nm and a gas flow rate of 50 cm3/min. The amplification is about 10 times that used in Fig. 7, (b) The response of the fibre optic sensing system when subjected to different 02 concentrations. (c) The response of the fibre optic sensing system with 02. (i) Plot of detector output against [O2]. (ii) Plot of l o g ( l N , / l o : ) against [02] using the data in (b).
for both O2 a n d CO2. The solvent mixture has a C C T A spectrum extending well into the visible region so it can be used with inexpensive plastic optical fibres. W a v e l e n g t h s at 400 a n d 520 n m were chosen to m o n i t o r the c o n c e n t r a t i o n s of 02 a n d CO2. A l t h o u g h plastic fibres have
substantial a t t e n u a t i o n a r o u n d 400 nm. the experimental set-up m e n t i o n e d above is still sensitive e n o u g h for 02 detection. Fluorescence m e a s u r e m e n t of F L 2- ions, in principle, can be a d a p t e d to detect C O , . However, we f o u n d that there was a cross-interference on CO, from 02
490
MING FAT CHoI and PETER HAWKINS
owing to the fluorescent quenching effect of molecular 02 on FL 2 ions. In addition, the determinations of 02 and CO2 can be accomplished easily by only switching between the monitoring wavelengths. Thus, all the
measurements were based on the absorption spectroscopic technique. Using the experimental set-up described earlier, the response, reproducibility and total signal change of the sensing system monitoring at 400 nm and a gas flow rate
(a)
~
(b)
CO 2 ,
E
8 o"
i
i r i,~,!
N2
i¸
o~
:'i Ii~
,
"2 :~[ii:r ::i j
~,
i!
i .... ,
i
},
i
;~i i
Time
:' q ,
i
L~
I
i
,
(min)
~
iil ~ . . . .
(c)
0.8
0.7 >~ 0.50'6
i
i,
0.4 0.3 0.2 0.1~
0
'
2
t
4
I
ICOd / %
6
8
10
Fig. ?. (a) The response time, reproducibility and total signal change of the fibre optic sensing system when subjected to changes between: N2--~ 10% CO 2 in N2--~ O2--, 10% CO 2 in N2--~ N 2. A solution (l.0 m]) of 10.6/zM FL and 190 # M K O H in 1:1 (v/v) D E A / D M F solvent mixture in a I cm cuvene was used with a wavelength of 520 nm and a gas flow rate of 50 cm3/min. (b) The response of the fibre optic sensing system when subjected to different CO 2 concentrations. (c) Plot of detector output against [CO2].
0_~/CO~-sensitive optical transducer
of 50 cm3/min through the solvent-dye solution in a 1 cm path-length cuvette were investigated. Figure 6a shows the response of the sensing system to step changes in gas concentrations from 100% N 2 to 100% 02, from 100% 02 to 10% CO2 in N 2, from 10% CO2 in N2 to 100% 02 and again back to 100% N 2. The reversibility of the sensing system is good and there is no cross-interference from CO2. The response time is 0.75 and 1.5 min for a 90% signal change from N 2 t o 02 and from 02 t o N 2 , respectively. The response of the sensing system when subjected to different levels of 02 was investigated (Fig. 6b). The decrease in signal level with increasing 02 concentrations is non-linear [Fig. 6c(i)] and the sensitivity (indicated by the slope of the graph) decreases as the concentration o f O 2 increases with about a 35% change of signal level from 0 to 25% 02. The sensing system responds over the range of 0-100% 02 with a response which is in agreement with Beer-Lambert's law so that a linear graph [Fig. 6c(ii)] is obtained when log(IN2/Io2) is plotted against [02] where IN, is the signal level recorded in N 2 only (i.e. [02] = 0) and lo2 is the level recorded in a gas mixture having an 02 concentration of [02]. The response, reproducibility and total signal change of the sensing system monitoring at 520 nm were investigated. Figure 7a shows the response of the sensing system to step changes in gas concentrations from 100% N 2 tO 10% C O 2 in N 2, from 10% CO2 in N 2 to 100% 02, from 100% 02 to 10% CO2 in N2 and again back to 100% N~. The reversibility of the sensing system is good and there is no cross-interference from O~. The response time is 0.41 min and 10.0 min for a 90% signal change from N 2 t o 1 0 % C O 2 in N~ and from 10% CO~ in N2 to N 2, respectively. The response of the sensing system when subjected to different concentrations of CO2 was investigated (Fig. 7b) and was shown to be reversible. The increase in signal level with increasing COt concentrations is non-linear (Fig. 7c): as a result, a non-linear calibration graph has to be plotted when using the sensing system for CO~ determination over the range of O- 10% CO~.
CONCLUSIONS
This study demonstrated the use of a single solvent~tye solution as an 02 and CO:-sensitive transducer for the determination of both gaseous 02 and CO2. This detecting system has
491
several advantages: inexpensive plastic optical fibre can be used; DEA and FL are inexpensive and readily available; preparation of solvent~lye solution is simple and fast. Possible applications include environmental and physiological monitoring of 02 over the ranges of 0 100% and CO2, 0-10%. The sensing medium however, still suffers from several drawbacks such as photo-bleaching of the dye, possibly cross-interference from other acidic or alkaline gases and also other electron acceptors like nitric oxide. 2s Furthermore, we expect the response of the solvent-dye solution to CO2 will be humidity dependent because of the aforementioned reasons. The presence of water is crucial for the response of the solvent~tye solution to CO~ and thus, increasing the water concentration in the solvent~lye solution is likely to shift the equilibrium of equation (2) from right to left and lowering the pH of the solvent-dye solution with a concomittant effect on the equilibrium of equation (3). The water content will certainly vary with the water vapour concentration in an applied gas mixture since D M F is a hydrophilic compound with a propensity to absorb water vapour. Secondly, we have observed that there is a hypsochromic shift of the absorbance band of FL 2 in D M F when the water content increases. 29 In order to circumvent this effect, the applied gas mixture can pass through some drying agent such as a molecular sieve before bubbling through the solvent dye solution. In addition, the sensing schemes are based on the concentration of dissolved O2 and CO, in the solvent~lye solution. Any variation of temperature will change the Henry's law constant of these gases and the solvent dye solution. Although we do not have the gas solubility data of these gases in D M T / D M F solvent mixture, it has been reported that there is a 11.6% decrease in solubility of CO, in aniline 3° and the solubilities of 02 in carbon tetrachloride, chlorobenzene, benzene and acetone increase by 2.00, 2.79, 4.35 and 4.02%, respectively, from 20 to 30~'C.3~ Consequently, we expect the response of the solvent~lye solution to CO, will be strongly dependent on temperature with a less marked dependence for 02. Finally, we are now planning to design a O2/CO2 fibre optic sensor by employing a similar design of Kar and Arnold? -~ Briefly, the solvent~lye solution is entrapped inside a segment of thin-wall poly(tetrafluoroethylene) (PTFE) tubing. A single optical fibre directs the
492
MING FAT CHOI and PETERHAWKINS
incident light into the solvent~lye solution through one end of the tubing and the transmitted light is collected by another optical fibre through the other end. Oxygen and CO2 gases can readily diffuse through the thin-wall PTFE tubing into the solvent~lye solution for the reactions. Acknowledgements--The authors would like to express their thanks to Dr A. Tubb for using his oxygen and carbon dioxide detectors and to the University of the West of England, Bristol for providing the financial support.
REFERENCES 1. M. L. Hitchman, Measurement of Dissoh,ed Oxygen, pp. 132-138, John Wiley & Sons, New York, 1978. 2. J. O. Lay, Metallurgia, 1955, 51, 109. 3. R. E. Ellis and B. Schurin, Appl. Optics, 1969, 8, 2265. 4. J. W. Severinghaus, Ann. N. Y. Acad. Sci., 1968, 148, 115. 5. J. I. Peterson, R. V. Fitzgerald and D. K. Buckhold, Anal. Chem., 1984, 56, 62. 6. A. Sharma and O. S. Wolfbeis, Appl. Spectrosc., 1988, 42, 1009. 7. P. Y. F. Li and R. Narayanaswamy, Analyst, 1989, 114, 1191. 8. C. Preininger, I. Klimant and O. S. Wolfbeis, Anal. Chem., 1994, 66, 1841. 9. G. G. Vurek, P. J. Feustel and J. W. Severinghaus, Ann. Biomed. Engng, 1983, 11, 499. 10. Z. Zhujun and W. R Seitz, Anal. Chim. Acta, 1984, 160, 305. 11. C. Munkholm, D. R. Walt and F. P. Milanovich, Talanta, 1988, 35, 109. 12. C. Goyet, D. R. Walt and P. G. Brewer, Deep-Sea Res., 1992, 39, 1015.
13. D. R. Walt, G. Gabor and C. Goyet, Anal. Chim. Acta, 1993, 274, 47. 14. M. D. DeGrandpre, Anal. Chem., 1993, 65, 331. 15. W. J. Albery and P. Barron. J. Electroanal. Chem. lnterfacial Electrochem., 1982, 138, 79. 16. Ph. Arquint, A. van den Berg, B. H. van der Schoot, N. F. de Rooij, H. B/ihler, W. E. Morf and L. F. J. Diirselen, Sens. Actuators B, 1993, 13-14, 340. 17. W. Gumbrecht, D. Peters, W. Schelter, W. Erhardt, J. Henke, J. Steil and U. Sykora, Sens. Actuators B, 1994, 18-19, 704. 18. J. L. Gehrich, D. W. Liibbers, N. Opitz, D. R. Hansmann, W. W. Miller, J. K. Tusa and M. Yafuso, IEEE Trans. Biomed, Engng, 1986, 33, 117. 19. W. W. Miller, M. Yafuso, C. F. Yan, H. K. Hui and S Arick, Clin. Chem. (Winston-Salem, N.C.), 1987, 33, 1538. 20. O. S. Wolfbeis, L. J. Weis, M. J. P. Leiner and W. E. Ziegler, Anal. Chem., 1988, 60, 2028. 21. D. F. Evans, J. Chem. Soc., 1953, 345. 22. H. Tsubomura and R. S. Mulliken, J. Am. Chem. Soc., 1960, 82, 5966. 23. A. U. Munck and J. F. Scott, Nature, 1956, 177, 587. 24. Y. Kawabata, T. Kamichika, T. Imasaka and N. lshibashi, Anal. Chim. Acta, 1989, 219, 223. 25. A. I, Vogel, Quantitative Inorganic Analysis, p. 243. Longmans, Green, London, 1961. 26. H. Diehl, Talanta, 1989, 36, 413. 27. N. O. Mchedlov-Petrossyan, M. I. Rubtsov and L. L. Lukatskaya, Dyes Pigments, 1992, 18, 179. 28. J. Jortner and U. Sokolov, J. Phys. Chem., 1961, 65, 1633. 29. M. F. Choi and P. Hawkins, Spectrosc. Lett., in press. 30. H. Stephen and T. Stephen (eds), Solubilities of Inorganic' Compounds, Vol. I, Part 2, p. 1068. Pergamon Press, Oxford, 1963. 31. W. F. Linke, Solubilities: Inorganic and Metal Organic Compounds, 4th Ed., Vol. 2, p. 1234. American Chemical Society, Washington, D.C., 1965. 32. S. Kar and M. A. Arnold, Talanta, 1994, 41, 1051.
Pergamon
0039-9140(95)01436-5
Talanta, Vol. 42, No. 3, pp. 483~,92, 1995 Copyright © 1995 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0039-9140/95 $9.50 + 0.00
A NOVEL OXYGEN AND/OR CARBON DIOXIDE-SENSITIVE OPTICAL TRANSDUCER MING FAT CHOI* a n d PETER HAWKINS Faculty of Applied Sciences, University of the West of England, Coldharbour Lane, Frenchay, Bristol BSI6 IQY, U.K.
(Receh~ed 2 September 1994. Accepted 12 September 1994)
Summary A novel oxygen (02) and/or carbon dioxide (CO 2)-sensitive transducer for the measurement of both gaseous 02 and CO., over the concentration ranges of O~, 0 100'% and CO,, 0 10% has been described employing a solution of 10.6 llM fluorescein (FL) and 190 FLMpotassium hydroxide in a solvent mixtures of 1:1 (v/v) N,N'-diethylaniline (DEA) and N,N-dimethylformamide. Increasing 02 concentrations cause the absorbance of the solution at a wavelength of 400 nm to increase owing to a contact charge transfer reaction existing between 02 and DEA molecules, and increasing CO~ concentrations produce a non-linear fall in absorbance at 520 nm as the colour of FL changes from its orange dianion form to the colourless neutral, lactonic form. Both processes are independent of each other and reversible. The response to changes in 02 concentrations is in good agreement with Beer Lambert's law and the response to changes in CO 2 concentrations in non-linear. A fibre optic sensing system based on this solvent-
O x y g e n (O2) can be considered to be one o f the m o s t i m p o r t a n t elements in o u r n a t u r e with an e n o r m o u s chemical a n d b i o c h e m i c a l reactions involving 02 as r e a c t a n t s or products. C a r b o n d i o x i d e (CO2) is usually a b y - p r o d u c t o f m a n y industrial processes a n d it is believed to aggravate the g r e e n h o u s e effect o f o u r e n v i r o n m e n t . M e d i c a l diagnosis a n d t r e a t m e n t o f critically ill patients in intensive care units a n d o p e r a t i n g theatres often requires m o n i t o r i n g o f CO2 and O: p a r t i a l pressures o f arterial b l o o d . Consequently, the d e t e r m i n a t i o n o f b o t h 02 a n d C O , is o f c o n s i d e r a b l e i m p o r t a n c e in e n v i r o n m e n t a l , b i o m e d i c a l analysis a n d analytical chemistry. C o n s i d e r a b l e effort has been d e v o t e d over m a n y years to the d e v e l o p m e n t o f new techniques for the m e a s u r e m e n t o f 02 a n d COz c o n c e n t r a t i o n s . A m p e r o m e t r i c a l d e t e r m i n a t i o n o f O, by a C l a r k - t y p e electrode ~ has been c o m m o n l y used in the past which suffers from the d r a w b a c k o f 02 c o n s u m p t i o n . The detection o f CO2 is usually based on an infrared d e t e c t o r 2"3 a n d a Severinghaus electrode 4 with b u l k y a n d expensive devices being used. W i t h the a d v e n t o f fibre optics technology, there is a g r o w i n g interest in the d e v e l o p m e n t o f fibre optics 02 a n d / o r CO2
sensors. The attractive features o f fibre optics sensors include: immunity to electrical interference; ease o f m i n i a t u r i z a t i o n ; c o m p a r a tively inexpensive device; r e m o t e a n d in situ sensing; a n d high i n f o r m a t i o n c a r r y i n g c a p a c i t y (multiwavelength transmission), etc. There have been m a n y p u b l i c a t i o n s on single optical fibre 02 sensors 5 8 based on the fluorescent q u e n c h i n g o f a dye by m o l e c u l a r 02 and C O , sensors 9 14 based on the p H m o d u l a t i o n a c c o m p l i s h i n g with the c o l o u r change o f a dye. Several electrochemical multi-sensors have also been r e p o r t e d for the s i m u l t a n e o u s l y detection o f both O, a n d CO,,. A l b e r y and B a r r o n ~5 a p p l i e d the electroreduction technique for the a m p e r o m e t r i c d e t e r m i n a t i o n o f 02 a n d CO2 in which two electrolytes and two a p p l i e d potentials were used. A r q u i n t et al. t6 a n d G u m b r e c h t et al. ~7 employed amperometric and potentiometric techniques for the detection o f pH, pO2 a n d pCO~. I n d i v i d u a l pH, pCO2 and pO2 sensors were fabricated and p o s i t i o n e d on a m i n i a t u r ized microchip. Recently, there have been a few c o m b i n e d fibre optic O~ and C O , sensors described in the literature. G e h r i c h et al. ~ a n d Miller et al. t9 a r r a n g e d three individual p H , pCO2 a n d pO2-sensitive optical fibres t o g e t h e r in a catheter. Similarly, oxygen-sensitive m a t e r i a l
*Author to whom correspondence should be addressed. tat
42 ~
t
483
484
MING FAT CHOl and PETERHAWKINS
and CO2-sensitive material were entrapped in separate gas-permeable membranes and attached to the distal end of a bifurcated optical fibre for sensing gaseous 02 and CO2.20 In all the above instances, individual sensitive-material was often used for each gas. To the best of our knowledge, not a single sensing medium for both 02 and CO2 have been reported to date. It has been known that some organic solvents exhibit contact charge transfer absorption (CCTA) bands when in contact with O2.2J Tsubomura and Mulliken22 explained that this is caused by the formation of CCT pairs between the organic solvents and molecular 02. The CCTA bands lie in longer wavelength regions if the solvents have lower ionization potentials, like some amine compounds, and usually appear as extensions to the long wavelength edge of the absorption spectra of the deoxygenated solvents. The CCTA intensity of the organic solvents is also related to the partial pressure of 02 in the gas mixtures above the solvents and the reaction between the solvents and 02 is reversiblefl 3 Fluorescein (FL) dye immobilized in poly(ethylene glycol) has been successfully applied for fibre optic sensing of CO2.24 With the combination of organic solvent and dye, a single transducer can possibly be employed for the determination of both 02 and CO2. In this paper, we apply this approach and report here a single 02/C02-sensitive transducer in which both gaseous 02 and CO2 can be detected independently based on the CCTA of N,N '-diethylaniline (DEA) with 02 and the colour change of FL upon exposure to CO2. EXPERIMENTAL
Chemicals and reagents DEA (>99%), N,N-dimethylformamide (DMF > 99.9%, HPLC grade), n-heptane (99%, spectrophotometric grade), methanol (>99.9%, HPLC grade) and FL dye (98%) were purchased from Aldrich Chemical Co. Ltd, U.K. Potassium hydroxide (KOH, AnalaR grade) was obtained from BDH Chemical Ltd, U.K. Solutions of 10.6 /~M FL and 190 /~M KOH in DMF, and 10.6 pM FL and 190 /~M KOH in 1:1 (v/v) DEA/DMF solvent mixture were prepared by adding 20 ~1 of a concentrated methanolic KOH solution (KOH concentration was determined by standard acid-base titration 25) and 70 /~1 of a concentrated FL in DMF solution (concentration of FL solution
was determined spectrophotometrically by employing the molar absorptivity of FL 2dianion 26) into DMF and !:1 (v/v) DEA/DMF solvent mixture, respectively. Similarly, a solution of 0.10 #M FL and 190/~M KOH in 1:1 (v/v) DEA/DMF solvent mixture was prepared for fluorescence studies. All reagents were used as received. Nitrogen gas ( S 2 ) , 0 2 gas, 10% (v/v) CO2 in N2 gas mixture and CO2 gas were supplied by Distillers MG, U.K.
The gas mixing system Different 02 (in the range of 0-100%) or CO2 concentrations (in the range of 0-10%) in a gas stream were produced by controlling the flow rates of either O2 gas or 10% CO2 in N 2 gas mixture and the diluent N 2 gas entering a mixing chamber. The gas mixture was passed through a portable 02 meter (Oxywarn 100I from Draeger Manufacturing, U.K.) or CO2 detector (LFG 10 Landfill Gas Analyser from Analytical Development Company Ltd, U.K.) where the 02 or CO2 concentrations in the gas mixture were determined before bubbling through the investigating solvent or solvent-dye solution contained in a 1 cm path-length quartz cuvette.
Instrumentation The CCTA spectra were measured on a Perkin-Elmer Lambda 15 Spectrophotometer, fluorescence excitation (EX) and emission (EM) spectra were recorded on a Perkin-Eimer LS-5 Luminescence Spectrometer. In the fibre optic detecting system of O2 and CO,,, a home-made optical arrangement was set up. A modulated beam of light with known wavelength was produced by a 100 W quartz halogen lamp (Bentham Instruments Ltd, U.K.) stabilized by a Bentham 505 current stabilized filament lamp power supply, a Bentham 218 optical chopper at a frequency of 141 Hz and a Bentham M300 monochromator controlled by a Bentham SMD3B stepping motor drive unit before being launched into a plastic optical fibre (I m long and 2 mm diameter from RS Components, U.K.). The transmitted light from the fibre irradiated the front surface of a I cm quartz cuvette containing 1 ml of solvent or solvent~zlye solution with gaseous O~ or CO, standards at a flow rate of about 50 cm3/min continuously passing through it. A second plastic optical fibre (0.5 m long, 2 mm diameter) was positioned on the back surface of the cuvette to collect the transmitted light. The other end of this fibre led directly onto the front surface of a
02/CO2-sensitive optical transducer
485
CCTA spectra for different mol percentage (%) mixtures of DEA/n-heptane mixture were also recorded (Fig. 2). There is an increase in absorbance as the mol% of DEA in DEA/nheptane mixture increases, n-Heptane has no CCTA band at the wavelengths investigated. A graph of abs at 400 nm against mol% of DEA in DEA/n-heptane mixture is plotted and a linear straight line is found which indicates that the CCTA is directly proportional to the concentration of DEA in DEA/n-heptane mixture. The change of absorbance arises from the CCTA of the DEA and 02 complex and the reaction is reversible.
100 mm 2 silicon photovoltaic detector (RS Components, U.K.) and the output signal was amplified by a Bentham 277 current preamplifier and a Bentham 223 lock-in amplifier which was synchronized to the modulated light beam by a reference signal provided by the light chopper. The amplified signal was recorded on a Linses IS chart recorder (Electroplan, U.K.) and displayed on a LED display (Bentham 217 digital unit). To avoid a slight drift caused by a small evaporation of the solvent, the solvent-dye solution was replaced with a fresh one at approximately 2 hr intervals. RESULTS AND DISCUSSION
hv
DEA +O2 ~ D E A ' • • 02 ~ ( D E A - ' • O2)*.
Development of O:/C02-sensitive transducer N,N'-Diethylaniline has a strong broad
It is possible that DEA can serve as a sensing medium for O2. It has also been observed that some pH-sensirive dyes change colour upon exposure to gaseous CO2. The presence of water is crucial for response to CO2 which is based on the following equations:
CCTA band lying in the ultraviolet/visible regions and there is an increase in absorbance as the 02 concentration increases (Fig. 1). The CCTA band disappears when the dissolved 02 is removed by purging DEA with N 2 or CO2. Note that the reference is air-saturated DEA ([02] = 21% approx.) which explains why DEA has negative absorbances when their 02 concentrations are less than the air-saturated value. A graph of absorbance (abs) at 400 nm against applied [02] is plotted where abs = absorbance of DEA saturated with a given applied [02] - absorbance of DEA saturated with N2. A linear straight line is obtained which agrees with Beer-Lambert's law having absorbance directly proportional to the applied [02]. Similarly, the
C O
2 -b
H20 ~ H2CO s
H z C O 3 + H 2 0 ~- HCO3 + H3 O+
H C O ; + H 2 0 ~ CO~- + H 3 0 + DH + H 2 0 ~ D - + H3 O+ , (colour A)
(colour B)
where D H and D - are the protonated and deprotonated forms of the dye, respectively. The
Abs 08
Slope n 0.04xi0' Intercept = 5.88xi0 ~ Correlation coefficient
06
Abs
= 0.9998
02
1.1 0
0
'
J ~0
4(I
'
~
GO
'
BO
[o21 / %
0.7
'
I00
1 : i 0 0 % 02 2 : 8 0 . 6 % O~ i n 3 : 6 3 . 3 % 02 in 4 : 4 5 . 3 % O~ i n 5 : air-saturated 6 : 1 2 . 3 % 02 in 7 : N~
0.3
N~ N~ N2 N2
k0.1
o,5
I
!
I
!
!
!
350
390
430
470
510
55O
Wavelength
(rim)
Fig. I. The effect of different 0 2 concentration on the C C T A spectrum of D E A . Inset shows the calibration plot of absorbance at 400 n m against [O2] using the data of the spectra. The optical pathlength is I cm
and the reference is air-saturated DEA.
486
M I N G FAT CHOI a n d PETER HAWKINS
Abs
06
Intercept ~ 2.15xi0 z Correlati~ coefficient = 0.9977
04
Abe
02
1.1
~ M o l e 2
%of DEAin D~n-hepmneMi~ures
0.7 3
0~9enated
100 78.6 58.0 38.0 1827
4
0.3
mole % of D E A
Deoxgenated
in D~/n-heptane mixtures ..............................................
~
5
1 2 3 4 5 6
u u
i' 2' 3' 4' 5" 6'
-0.1
4' 5" -0.5
F h ' 350
!
I
|
!
!
400
450
500
550
600
Wavelength
(nm)
Fig. 2. The CCTA spectra o f D E A / n - h e p t a n e solvent mixtures. Solvent mixtures are in mol%. Inset s h o w s the calibration plot o f absorbance at 400 nm against m o l % o f D E A in D E A / n - h e p t a n e solvent mixture. The measurement conditions are the same as in Fig. I.
colour change of the dye is sensitive to pH change of its environment. A solution of 10.6 /tM FL and 190 # M KOH in D M F was tested for its response to CO2 (Fig. 3). It is assumed that FL exists only as FL 2 ions in this highly alkaline solvent. Neutral form of FL can occur in three different tautomers, i.e. zwitterion, quinoid and lactone. However, it has been reported that colourless lactonic form is usually the dominant tautomer present in organic solvents. 27 The change of absorbance arises from the conversion of the orange dianion form of FL to the colourless, neutral lactonic form upon exposure to CO2. These changes were brought
about from the change of the pH of the solution with the dissolution of CO2 in D M F and can be envisaged in the following reactions: CO2+OH
~ HCO3
(I)
~ H20
(2)
H + + OH
2H + + FL 2- ~ H2FL.
(3)
Dissolved CO2 removed the O H - ions in DMF: as a result, the H + ion concentration is increased which shifts the equilibrium of equation (3) from left to right. A plot of abs at 520 nm against [CO2] is shown in Fig. 3 and increasing [CO2] cause a non-linear fall in
Abs /kbs
1. 28" 04
0.?2' 0"20
2
4
6
8
10
[co2] / % 0.64'
/
/
3
¢
1 : Nl 2 : 0.88 3 : 1.99 4 : 2.78 5 : 3.34 6 : 4.60 7 : 6.20 8 : 8.73 9 : 10.0
i0:
i00
% % % % % % % %
CO 2 CO 2 CO 2 CO~ CO 2 CO 2 CO~ CO~
in in in in in in in in
N2 N2 N2 N~ N3 N2 N~ Nz % CO 2 in N 2
8.36
8.88' 10
- 0 . 2 e ..................................................... •........................................................ , .......................................................... ' ....................................................... ' .........................................' 388 424 46~ 512 556 600 Wavelength
(ran)
Fig. 3. The effect o f different C O , concentration on the absorption spectrum o f a solution o f 10.6 p M FL and 190 p M K O H in D M F . Inset s h o w s the plot o f abs against [CO2] using the data o f the spectra. T h e optical pathlength is 1 cm and the reference is a solution o f air-saturated 190 ~tM KOH in DMF.
O,/C02-sensitive optical transducer (a)
487
Abs
1 .20
092"
] I
O 64
1
:
100
2 3 4 5 6 7 8 9 i0 Ii
: : : : : : : : : :
87.0 % 02 in 7 1 . 6 % 0 z in 55.9 % 02 in 4 2 . 0 % O 2 in 3 0 . 0 % O 2 in 26.3 % O~ in air-saturated 14.6 % O~ in 7.5 % o~ in N2
% % N2 N2 N2 N2 N2 N1 N2 N2
12
12 13 14 15 16 17 18 19 20 21 22
: : : : : : : : : : :
0.38 0.70 1.35 2.44 3.03 3.82 5.14 6.34 8.39 10.0 100
% % % % % % % % % % %
C02 C% CO 2 C02 CO s CO 2 C02 CO 2 COz CO 2 CO:
in in in in in in in in in in
N2 N2 N2 N2 N2 N2 N~ N2 N2 N~
0
0.08
16 17 18 19 20
-0.2e
424
380
468
21
22
51~ Wavelength
556
!
600
(nm)
[Oz] / %
(b) 0
20 '
I
40 '
I
60 I
'
,
80 ,
,
I00 ~'110'4
ill)
0.2
0
2
4
6
8
10
[CO2I / %
Fig. 4. (a) The effect of 02 and CO2 on the absorption spectrum of a solution of 10.6/zM FL and 190 FM KOH in 1: 1 (v/v) DEA/DMF solvent mixture. Optical pathlength: 1 cm; reference: air-saturated 190 FM KOH in l:l (v/v) DEA/DMF solvent mixture. (b)(i) Plot of abs against [02] at 400 nm using the data in (a). (ii) Plot of abs against [CO2] at 520 nm using the data in (a).
absorbance with a 91% change in absorbance from 0 to 10% CO2. We observed that a solution of FL in D E A changes colour on exposure to CO2 but that the change is not entirely reversible which may be probably owing to DEA being an almost completely non-aqueous solvent. As mentioned above, a solution of FL in D M F responds reversibly to gaseous CO2 as D M F contains a small amount of dissolved water (about 0.03%). Unfortunately, D M F has a CCTA spectrum which is too far into the ultraviolet region to be of practical use for future fibre optic sensor development. After further investigation, we
devised a way around this problem by using a solution of FL dye and KOH in a 1:i (v/v) D E A / D M F solvent mixture for the independent and reversible measurements of CO2 (using colour changes of the FL dye) and 02 (using the CCTA of the D E A with 02). The observed changes in absorption spectrum of the solventdye solution as a function of [02] and [CO,,] are shown in Fig. 4a which demonstrates that there are no cross-interference between the two gases. A graph of abs at 400 nm against applied [02] is plotted in Fig. 4b(i). The results agree with Beer-Lamberrs law and the absorbance is linearly proportional to the applied [02]. The
488
MING FAT CHOI and PETERHAWKINS
(a)
(b)
(c)
# 6e--
1 2 3 4 5 6 7 8
: : : : : : : :
9 : 10 : Ii : 12 : 13 14 : 15 :
48
:
"~
air-saturated 0.07 0.19 0.32 0.44 0.52 0.76
% % % % % %
CO~ CO2 CO2 CO~ C02 CO2
in in in in in in
0.99 1.79 3.22 5.06 6.44
% % % % %
CO~ CO~ C~ C02 CO~
in N 2 in N 2 N~ in N 2 in l~
N2 N~ N2 N2 N~
in
]
i0.0 % CO~ in N~ I00
6e-
12 3
N2
3 s
t CO~
"
i
~
7
~
89
-
lO
7
20
1 3 89
1~21_3~.,is I
4~O
448
400
L-
~20
Wavelength
I
--
-I
~60
1231415
8
~2e
608
(ran)
~O
Wavelength
600
i0111213 i~15
e 520
(nm)
560
Wavelength
608 (nm)
(d)
Intensity 60 _
-
. . . . . . . . . . .
° - ~
- - = _ _
50 40 -": 30
20 10 r~.___~
0o
2
T--
-.-~_._ ~ .......
4
~. . . . . . . . . . .
6
.
,
8
[C021 / % Fig. 5. Fluorescence excitation (EX) and emission (EM) spectra of 0.10/aM FL and 190 #M KOH in 1:1 (v/v) DEA/DMF solvent mixture. Optical pathlength:l cm. (a) EX spectra with EM wavelength at 538 nm. (b) EM spectra with EX wavelength at 520 nm. (c) EM spectra with EX wavelength at 490 nm. (d) Plot of intensity against applied [CO2] using the data of the spectra.
change of absorbance is again caused by the CCTA of the DEA and 02 complex. A plot of abs at 520 nm against applied [CO2] is shown in Fig. 4b(ii) and increasing [CO2] produce a nonlinear fall in absorbance. The change of absorbance arises from the conversion of the orange dianion form of FL to the colourless, neutral lactonic form upon exposure to CO2. The fluorescence EX and EM spectra of 0.10 /~M FL and 190 /tM KOH in 1:1 (v/v) D E A / D M F solvent mixture is shown in Figs. 5a-c. The fluorescence intensity is decreased when the CO2 concentration is increased. Unlike FL 2- ion which is a strongly
fluorescent molecule, the neutral lactonic form of FL is non-fluorescent. As explained above, FL 2- ions in the solution are probably converted to neutral lactonic form of FL upon exposure to CO2 (equations I-3). The plot of intensity against applied [CO2] is shown in Fig. 5d which shows that increasing [CO2] generates a non-linear fall in fluorescence intensity.
Fibre optic sensing system for 02 and CO_, The experimental results demonstrate that FL dye and KOH in D E A / D M F solvent mixture is a promising 02 and CO2-sensitive transducer to be used for the development of a sensing system
O2/CO2-sensitive optical transducer
489
(a) 10% COy
(b)
::
o o
,n
Time
(min)
(c) 0.7 slope. 4 a~xlO' Intercept - 4 4xze ~
,,
0.6
:o..,,,
....
-
,,o, ....
o-,,
.-
>~ o
"~0.4
....
0.3 0.5
~;~.-
S
"
."
0,2 <
0,4
0.3 .
t
0
0.1
,:. '~
L
20
40
60
1~ J O
80
100
[Od / %
Fig. 6. (a) The response time, reproducibility and total signal change of the fibre optic sensing system when subjected to changes between: N2---*O2---~10% CO2 in N2---.O2---~N2. 1.0 ml of a solution of 10.6 #M FL and 190 /~M KOH in 1:1 (v/v) DEA/DMF solvent mixture in a 1 cm cuvette was used with a wavelength of 400 nm and a gas flow rate of 50 cm3/min. The amplification is about 10 times that used in Fig. 7, (b) The response of the fibre optic sensing system when subjected to different 02 concentrations. (c) The response of the fibre optic sensing system with 02. (i) Plot of detector output against [O2]. (ii) Plot of l o g ( l N , / l o : ) against [02] using the data in (b).
for both O2 a n d CO2. The solvent mixture has a C C T A spectrum extending well into the visible region so it can be used with inexpensive plastic optical fibres. W a v e l e n g t h s at 400 a n d 520 n m were chosen to m o n i t o r the c o n c e n t r a t i o n s of 02 a n d CO2. A l t h o u g h plastic fibres have
substantial a t t e n u a t i o n a r o u n d 400 nm. the experimental set-up m e n t i o n e d above is still sensitive e n o u g h for 02 detection. Fluorescence m e a s u r e m e n t of F L 2- ions, in principle, can be a d a p t e d to detect C O , . However, we f o u n d that there was a cross-interference on CO, from 02
490
MING FAT CHoI and PETER HAWKINS
owing to the fluorescent quenching effect of molecular 02 on FL 2 ions. In addition, the determinations of 02 and CO2 can be accomplished easily by only switching between the monitoring wavelengths. Thus, all the
measurements were based on the absorption spectroscopic technique. Using the experimental set-up described earlier, the response, reproducibility and total signal change of the sensing system monitoring at 400 nm and a gas flow rate
(a)
~
(b)
CO 2 ,
E
8 o"
i
i r i,~,!
N2
i¸
o~
:'i Ii~
,
"2 :~[ii:r ::i j
~,
i!
i .... ,
i
},
i
;~i i
Time
:' q ,
i
L~
I
i
,
(min)
~
iil ~ . . . .
(c)
0.8
0.7 >~ 0.50'6
i
i,
0.4 0.3 0.2 0.1~
0
'
2
t
4
I
ICOd / %
6
8
10
Fig. ?. (a) The response time, reproducibility and total signal change of the fibre optic sensing system when subjected to changes between: N2--~ 10% CO 2 in N2--~ O2--, 10% CO 2 in N2--~ N 2. A solution (l.0 m]) of 10.6/zM FL and 190 # M K O H in 1:1 (v/v) D E A / D M F solvent mixture in a I cm cuvene was used with a wavelength of 520 nm and a gas flow rate of 50 cm3/min. (b) The response of the fibre optic sensing system when subjected to different CO 2 concentrations. (c) Plot of detector output against [CO2].
0_~/CO~-sensitive optical transducer
of 50 cm3/min through the solvent-dye solution in a 1 cm path-length cuvette were investigated. Figure 6a shows the response of the sensing system to step changes in gas concentrations from 100% N 2 to 100% 02, from 100% 02 to 10% CO2 in N 2, from 10% CO2 in N2 to 100% 02 and again back to 100% N 2. The reversibility of the sensing system is good and there is no cross-interference from CO2. The response time is 0.75 and 1.5 min for a 90% signal change from N 2 t o 02 and from 02 t o N 2 , respectively. The response of the sensing system when subjected to different levels of 02 was investigated (Fig. 6b). The decrease in signal level with increasing 02 concentrations is non-linear [Fig. 6c(i)] and the sensitivity (indicated by the slope of the graph) decreases as the concentration o f O 2 increases with about a 35% change of signal level from 0 to 25% 02. The sensing system responds over the range of 0-100% 02 with a response which is in agreement with Beer-Lambert's law so that a linear graph [Fig. 6c(ii)] is obtained when log(IN2/Io2) is plotted against [02] where IN, is the signal level recorded in N 2 only (i.e. [02] = 0) and lo2 is the level recorded in a gas mixture having an 02 concentration of [02]. The response, reproducibility and total signal change of the sensing system monitoring at 520 nm were investigated. Figure 7a shows the response of the sensing system to step changes in gas concentrations from 100% N 2 tO 10% C O 2 in N 2, from 10% CO2 in N 2 to 100% 02, from 100% 02 to 10% CO2 in N2 and again back to 100% N~. The reversibility of the sensing system is good and there is no cross-interference from O~. The response time is 0.41 min and 10.0 min for a 90% signal change from N 2 t o 1 0 % C O 2 in N~ and from 10% CO~ in N2 to N 2, respectively. The response of the sensing system when subjected to different concentrations of CO2 was investigated (Fig. 7b) and was shown to be reversible. The increase in signal level with increasing COt concentrations is non-linear (Fig. 7c): as a result, a non-linear calibration graph has to be plotted when using the sensing system for CO~ determination over the range of O- 10% CO~.
CONCLUSIONS
This study demonstrated the use of a single solvent~tye solution as an 02 and CO:-sensitive transducer for the determination of both gaseous 02 and CO2. This detecting system has
491
several advantages: inexpensive plastic optical fibre can be used; DEA and FL are inexpensive and readily available; preparation of solvent~lye solution is simple and fast. Possible applications include environmental and physiological monitoring of 02 over the ranges of 0 100% and CO2, 0-10%. The sensing medium however, still suffers from several drawbacks such as photo-bleaching of the dye, possibly cross-interference from other acidic or alkaline gases and also other electron acceptors like nitric oxide. 2s Furthermore, we expect the response of the solvent-dye solution to CO2 will be humidity dependent because of the aforementioned reasons. The presence of water is crucial for the response of the solvent~tye solution to CO~ and thus, increasing the water concentration in the solvent~lye solution is likely to shift the equilibrium of equation (2) from right to left and lowering the pH of the solvent-dye solution with a concomittant effect on the equilibrium of equation (3). The water content will certainly vary with the water vapour concentration in an applied gas mixture since D M F is a hydrophilic compound with a propensity to absorb water vapour. Secondly, we have observed that there is a hypsochromic shift of the absorbance band of FL 2 in D M F when the water content increases. 29 In order to circumvent this effect, the applied gas mixture can pass through some drying agent such as a molecular sieve before bubbling through the solvent dye solution. In addition, the sensing schemes are based on the concentration of dissolved O2 and CO, in the solvent~lye solution. Any variation of temperature will change the Henry's law constant of these gases and the solvent dye solution. Although we do not have the gas solubility data of these gases in D M T / D M F solvent mixture, it has been reported that there is a 11.6% decrease in solubility of CO, in aniline 3° and the solubilities of 02 in carbon tetrachloride, chlorobenzene, benzene and acetone increase by 2.00, 2.79, 4.35 and 4.02%, respectively, from 20 to 30~'C.3~ Consequently, we expect the response of the solvent~lye solution to CO, will be strongly dependent on temperature with a less marked dependence for 02. Finally, we are now planning to design a O2/CO2 fibre optic sensor by employing a similar design of Kar and Arnold? -~ Briefly, the solvent~lye solution is entrapped inside a segment of thin-wall poly(tetrafluoroethylene) (PTFE) tubing. A single optical fibre directs the
492
MING FAT CHOI and PETERHAWKINS
incident light into the solvent~lye solution through one end of the tubing and the transmitted light is collected by another optical fibre through the other end. Oxygen and CO2 gases can readily diffuse through the thin-wall PTFE tubing into the solvent~lye solution for the reactions. Acknowledgements--The authors would like to express their thanks to Dr A. Tubb for using his oxygen and carbon dioxide detectors and to the University of the West of England, Bristol for providing the financial support.
REFERENCES 1. M. L. Hitchman, Measurement of Dissoh,ed Oxygen, pp. 132-138, John Wiley & Sons, New York, 1978. 2. J. O. Lay, Metallurgia, 1955, 51, 109. 3. R. E. Ellis and B. Schurin, Appl. Optics, 1969, 8, 2265. 4. J. W. Severinghaus, Ann. N. Y. Acad. Sci., 1968, 148, 115. 5. J. I. Peterson, R. V. Fitzgerald and D. K. Buckhold, Anal. Chem., 1984, 56, 62. 6. A. Sharma and O. S. Wolfbeis, Appl. Spectrosc., 1988, 42, 1009. 7. P. Y. F. Li and R. Narayanaswamy, Analyst, 1989, 114, 1191. 8. C. Preininger, I. Klimant and O. S. Wolfbeis, Anal. Chem., 1994, 66, 1841. 9. G. G. Vurek, P. J. Feustel and J. W. Severinghaus, Ann. Biomed. Engng, 1983, 11, 499. 10. Z. Zhujun and W. R Seitz, Anal. Chim. Acta, 1984, 160, 305. 11. C. Munkholm, D. R. Walt and F. P. Milanovich, Talanta, 1988, 35, 109. 12. C. Goyet, D. R. Walt and P. G. Brewer, Deep-Sea Res., 1992, 39, 1015.
13. D. R. Walt, G. Gabor and C. Goyet, Anal. Chim. Acta, 1993, 274, 47. 14. M. D. DeGrandpre, Anal. Chem., 1993, 65, 331. 15. W. J. Albery and P. Barron. J. Electroanal. Chem. lnterfacial Electrochem., 1982, 138, 79. 16. Ph. Arquint, A. van den Berg, B. H. van der Schoot, N. F. de Rooij, H. B/ihler, W. E. Morf and L. F. J. Diirselen, Sens. Actuators B, 1993, 13-14, 340. 17. W. Gumbrecht, D. Peters, W. Schelter, W. Erhardt, J. Henke, J. Steil and U. Sykora, Sens. Actuators B, 1994, 18-19, 704. 18. J. L. Gehrich, D. W. Liibbers, N. Opitz, D. R. Hansmann, W. W. Miller, J. K. Tusa and M. Yafuso, IEEE Trans. Biomed, Engng, 1986, 33, 117. 19. W. W. Miller, M. Yafuso, C. F. Yan, H. K. Hui and S Arick, Clin. Chem. (Winston-Salem, N.C.), 1987, 33, 1538. 20. O. S. Wolfbeis, L. J. Weis, M. J. P. Leiner and W. E. Ziegler, Anal. Chem., 1988, 60, 2028. 21. D. F. Evans, J. Chem. Soc., 1953, 345. 22. H. Tsubomura and R. S. Mulliken, J. Am. Chem. Soc., 1960, 82, 5966. 23. A. U. Munck and J. F. Scott, Nature, 1956, 177, 587. 24. Y. Kawabata, T. Kamichika, T. Imasaka and N. lshibashi, Anal. Chim. Acta, 1989, 219, 223. 25. A. I, Vogel, Quantitative Inorganic Analysis, p. 243. Longmans, Green, London, 1961. 26. H. Diehl, Talanta, 1989, 36, 413. 27. N. O. Mchedlov-Petrossyan, M. I. Rubtsov and L. L. Lukatskaya, Dyes Pigments, 1992, 18, 179. 28. J. Jortner and U. Sokolov, J. Phys. Chem., 1961, 65, 1633. 29. M. F. Choi and P. Hawkins, Spectrosc. Lett., in press. 30. H. Stephen and T. Stephen (eds), Solubilities of Inorganic' Compounds, Vol. I, Part 2, p. 1068. Pergamon Press, Oxford, 1963. 31. W. F. Linke, Solubilities: Inorganic and Metal Organic Compounds, 4th Ed., Vol. 2, p. 1234. American Chemical Society, Washington, D.C., 1965. 32. S. Kar and M. A. Arnold, Talanta, 1994, 41, 1051.