United States	Office of Research	EPA/600/R-96/137
Environmental Protection	and Development	November 1996
Agency	Washington DC 20460
Detection of Phenols using
Liquid Chromatographic
System with an Enzyme-
based Biosensor
U2ASB96.RPT

-------
EPA/600/R-96/137
November 1996
Detection of Phenols using
Liquid Chromatographic System
with an Enzyme-based Biosensor
by
Joseph Wang
Department of Chemistry and Biochemistry
New Mexico State University
Las Cruces, NM 88003
(Purchase Order Number: 5V2154NATX)
Prepared for
Kim Rogers
National Exposure Research Laboratory
Characterization Research Division
P.O. Box 93478
Las Vegas, Nevada 89193-3478
United States Environmental Protection Agency
Office of Research and Development
National Exposure Research Laboratory
Characterization Research Division
P.O. Box 93478
Las Vegas, Nevada 89193-3478
142asb96.rpt

-------
Notice
The U.S. Environmental Protection Agency (EPA), through its Office of Research and Development
(ORD), has funded and managed the research described here. It has been subjected to the Agency's peer
review and has been approved as an EPA publication.
ii

-------
Foreword
One of the approaches for reducing uncertainties in the assessment of human exposure is to better
characterize the hazardous wastes which contaminate our environment. A significant limitation to this
approach, however, is that sampling and laboratory analysis of contaminated air, water, and soil, is slow and
expensive, thus limiting the number of samples which can be analyzed within time and budget constraints.
In cases where a limited number of target analytes can be identified, faster and more cost-effective field
screening and monitoring methods can potentially increase the amount of information available concerning
the location and concentration of pollutants which may impact human health and the environment.
Due primarily to their operational format versatility, biosensors composed of biological recognition
elements interfaced with optical, electrochemical, or acoustic signal transducers, show the potential to provide
solutions for the previously mentioned limitations. In particular, the enzyme electrodes may fill in some of
the gaps currently found in the field analytical technologies matrix. Some of the potential application areas
for which these biosensors might be developed include: in situ monitoring for on-line process control for
pollution prevention and remediation scenarios; sentinel capabilities for detection of episodic releases or
continuous well monitoring in hazardous waste site post-closure monitoring scenarios; and detectors for field
chromatographic systems which use detergent micelles or organic-based solvent systems.
The following research efforts, supported through the National Exposure Research Laboratory (NERL),
Characterization Research Division (CRD-LV), Analytical Sciences Branch (ASB), are intended to develop
and demonstrate electrochemical biosensors for detection of environmental pollutants such as phenolics. This
work is part of a broader effort in which biosensors are being developed to function in potential applications
which require portable, continuous, and in situ monitoring of aqueous and organic-based media.
iii

-------
Contents
Notice	 ii
Foreword 	iii
Figures	 v
Tables	 v
Chapter 1 Goal	 1
Chapter 2 Background	 2
Chapter 3 Experimental		3
Reagents 		3
High Performance Liquid Chromatography		3
Electrode Design and Preparation		3
Sample Analysis		3
Chapter 4 Results and Discussion	 5
Chapter 5 Conclusions and Recommendation	 12
References 	 13
iv

-------
Figures
3-1.	Schematic for (dual carbon paste electrode) electrochemical flow cell detector	4
4-1.	Preliminary results of phenol analysis using reverse phase HPLC-dual electrode 	6
4-2. Chromatograms of 15-component phenol mixture at pH 6.5, under
optimum experimental conditions 	8
4-3. Typical calibration plots for 2,4-dinitrophenol and phenol at tyrosinase (3.0%)
carbon paste electrode, using optimal experimental conditions	9
4-4. Typical calibration plots for p-nitrophenol and o-cresol at tyrosinase (3.0%)
carbon paste electrode, using optimal experimental conditions	10
4-5. Sample analyses. Typical chromatograms of real water samples spiked with a 12.5^M
four-component phenol mixture under optimum experimental conditions	11
Tables
4-1. Retention times, detection limits and peak ratios of a mixture of 15 phenolic compounds	7
v

-------
Chapter 1
Goal
The objective of this project is to characterize,
optimize and test a dual-electrode detector for
monitoring phenolic pollutants in liquid chromato-
graphic effluents. Particular attention is given to the
simultaneous use of a bioelectrode and conventional
electrodes for the detecting of phenolics.
1

-------
Chapter 2
Background
Because of the inherent toxicity of phenolic
compounds, there is considerable interest for their
determination in environmental matrices. Such
compounds are reported at over 80 Superfund sites,
and are listed on the Agency for Toxic Substances
and Disease Registry's (ATSDR's) Priority Haz-
ardous Compound List. Electrochemical sensors and
detectors are extremely attractive for monitoring
phenolic compounds. The phenol moiety can be
oxidized at moderate potentials (ca. +1.0V vs.
Ag/AgCl reference) hence opening the way for direct
anodic detection of phenols (1). In addition, the
enzymatic activity of tyrosinase (polyphenol oxidase,
EC 1.14.18.1) can be coupled with amperometric
transduction to yield effective biocatalytic sensors for
phenols. Such a biosensing operation relies on the
enzymatic conversion of phenols to quinones and
subsequent low-potential reductive detection of the
quinone product (2, 3).
This project employs both the direct- and
enzymatic amperometric schemes for monitoring
phenolic compounds in chromatographic effluents.
Reverse-phase liquid chromatography represents a
very useful tool for the separation of phenols. Dual-
electrode amperometric detection for HPLC can offer
additional information towards the identification of
eluting (and coeluting) compounds (3). Commonly,
such a detection scheme relies on the use of identical
working electrodes, operated at different potentials,
in connection with the parallel thin-layer cell
configuration. In this project, the same detector
configuration was employed, but in connection with
a tyrosinase electrode (held at -0.2V) and a plain
carbon electrode (operated at +1.2V). Such a unique
dual-electrode detection mode adds a new dimension
of information and selectivity, and hence greatly
enhances the high performance liquid chroma-
tography (HPLC) detection of phenolic pollutants.
2

-------
Chapter 3
Experimental
Reagents
Helium gas (Argyle Welding Supply,
Albuquerque, NM) was used as received for
degassing the mobile phase and other solutions.
Acetonitrile and methanol, both HPLC grade were
obtained from Sigma. 2,4-dinitrophenol, 2,4,6-
trichlorophenol, pentachlorophenol, 2,4-dichloro-
phenol, 2,4,5-trichlorophenol, 4-nitrophenol, p-
cresol, o-cresol, 2-chlorophenol, 2-nitrophenol, and
p-chlorophenol were all obtained from Sigma. 2,4-
dimethylphenol, creosol, 4,6-dinitro-o-cresol, were
from Aldrich, and phenol was from Fisher.
Tyrosinase enzyme (E.C. 1.14.18.1., T-7755),
4,400 Units/mg solid, was obtained from Sigma.
Mineral oil and graphite powder were obtained from
Aldrich and Fisher, respectively. Sodium phosphate
buffer (PB, 0.05M) was adjusted to pH 6.5. HPLC
grade water was used throughout the experiment.
Environmental water samples were obtained from:
Oak Ridge Creek Water, Tennessee, and Hanford
Ground Water, Washington State. Membrane filters
include: nylon membranes and microfilters, 0.2/im
pore size, 47mm. All electrodes and electroanalizers
were obtained from Bioanalytical Systems Inc.
High Performance Liquid Chromatography
All standard solutions were prepared in the mobile
phase. All mobile phases, samples and solutions
were filtered and degassed prior to use. The most
frequently used mobile phase was acetonitrile
(25:75) PB, pH 6.5, 0.05M. All experiments were
performed on a reverse phase high performance
liquid chromatography system (Model BAS 480,
Bioanalytical Systems Inc., Lafayette, IN). A dual
reciprocating pump PM-80 (BAS), complete with a
dual electrochemical detector LC-4C (BAS), was
used with BAS thin layer electrochemical cell in
parallel mode. A flow rate of 0.5ml/min was used
throughout the experiment. The plain carbon paste
electrode (CPE) was operated at +1.2V, and the
enzyme electrode, tyrosinase (3.0%) CPE, at -0.2V.
Injections were made manually through a 20/il
Reodyne loop injection valve, onto a BAS-C,,
column (BAS, MF6213), ODS-3, 3^m, 3.2x 100mm.
Chromatograms were recorded on an Omniscribe
strip chart dual pen recorder (Houston Instruments).
Electrode Design and Preparation
The thin layer cell (Figure 3-1) was operated in
parallel mode. One cavity contained the plain carbon
paste electrode (CPE) prepared using a 60:40
graphite powder: mineral oil composition, which was
mixed thoroughly, and then packed into the electrode
cavity. The second cavity was used for the enzyme
electrode and was prepared using carbon paste, with
a tyrosinase content of 3% (w/w). Both electrodes
were carefully polished using ultra high purity
weighing paper, taking care not to cross contaminate
the surfaces. The electrode cell was then placed into
its housing unit, the carrier buffer flow initiated, and
the appropriate potentials applied: +1.20V for the
plain CPE, and -0.20V for tyro-sinase (3.0%) CPE.
Current was monitored, using the strip chart reorder
and the baselines were allowed to stabilize before
any analysis was performed.
Sample Analysis
Quantities of water samples from Hanford Ground
Water (pH 7.8), WA, and Oak Ridge Creek Water
(pH 6.6), TN, were spiked with known
concentrations of phenol compounds. The samples
were filtered and degassed, and without further
pretreatment, injected onto the column. A control
was also performed by injecting unspiked water
samples. Calibration curves were also prepared, and
the percentage recovery for the samples determined.
3

-------
Outlet	Inlet
Figure 3-1. Schematic for (dual carbon paste electrode) electrochemical flow cell detector.
4

-------
Chapter 4
Results and Discussion
Preliminary experiments were conducted for
optimizing the enzyme-electrode and mobile-phase
compositions, as well as experimental variables such
as the flow rate and operating potentials. An enzyme
loading of 3%w (in the carbon paste) coupled to a
mobile phase containing 25% acetonitrile and 75%
phosphate buffer (0.05M, pH 6.5) were thus selected
for subsequent work, along with operating potentials
of -0.2 and +1.2V for the enzyme- and plain
electrodes, respectively.
Figure 4-1 displays the dual-electrode liquid
chromatography with electrochemical detection
(LCEC) response for a mixture containing 0.5mM
2,4-dinitrophenol, phenol, p-nitrophenol and o-cresol
(peaks 1, 2, 3, and 4, respectively). Both the enzyme
(A) and the plain (B) electrodes respond favorably,
with high signal-to-noise characteristics, to these
phenolic pollutants. (Notice the different current
scales.) These data demonstrate also the
compatibility of the tyrosinase electrode with the
partially organic mobile phase; the latter is expected
from the known organic-phase activity of tyrosinase
(4).
Table 4-1 summarizes the retention times,
detection limits and enzyme/plain-electrodes peak
ratios for 15 environmentally important phenolic
contaminants. Most species, with the exception of
2,4,5-trichlorophenol, are eluted within less than 25
minutes. The enzyme electrode detectors offer ex-
tremely low detection limits; these range from 1.18ng
(for phenol) to ca. 400ng (for pentachlorophenol).
The peak-current ratios (enzyme/bare electrodes)
range from 0.76 (for 4,6-dinitro-o-cresol) to 1.10 (for
2,4-dichlorophenol). Such different peak ratios
greatly enhances the information content of the
LCEC response. A typical dual-electrode response
for a mixture of these 15 phenolic pollutants is
displayed in Figure 4-2.
5

-------
Peak#
Compound
(0.5mM)
1
2,4-dinitrophenol
2
Phenol
3
p-nitrophenol
4
o-cresol
<
e
S
s
s
I
s
1
16mln
i

-------
EPA Project
Conditions
Mobile Phase:	25:75 Acetonitrile:PBS pH 6.5,0.05M
Flow Rate:	0.5ml/min
Column:	BASC-18
Loop Size\	20^1
Table 4-1. Retention times, detection limits and peak ratios of a mixture of 15
phenolic compounds.
Peak
Compound
Retention
Time (min)
Detection
Limit* (ng)
Peak
Ratios"
1
2,4-Dinitrophenol
1.92
6.90
0.78
2
4,6-Dinitro-o-cresol
2.40
6.19
0.76
3
Phenol
2.80
1.18
0.80
4
4-Nitrophenol
3.52
2.17
0.85
5
p-creso 1
6.00
6.76
0.84
6
o-cresol
6.80
3.38
0.86
7
creosol
6.80
3.05
0.80
8
2-Chlorophenol
7.28
2.19
0.85
9
2-Nitrophenol
7.44
8.70
0.85
10
p-Chlorophenol
9.40
2.17
0.88
11
2,4-Dimethylphenol
11.80
9.16
0.85
12
2,4,6-T richlorophenol
16.40
98.70
0.95
13
Pentachlorophenol
18.10
399.45
0.97
14
2,4-Dichlorophonol
22.40
3.44
1.10
15
2,4,5-T richlorophenol
46.20
74.03
0.97
* Detection limits were obtained at signal-to-noise of 3, and calculated for tyrosinase
(3.0%) CPE only.
** Peak ratios calculated,	^ryL.)'1Q.
Figures 4-3 and 4-4 display calibration plots
obtained at the tyrosinase-based flow detector for
2,4-dinitrophenol, phenol, p-nitrophenol and o-
cresol. All four phenolic compounds display good
linearity over the entire concentration range (0 •
25//M) examined. The slopes of these plots (i.e., the
sensitivity) correspond to O.381-(o-cresol), 0.424-(p-
nitrophenol), 0.449-(2,4-
-------
Peak#
Compound
1
2,4-Dinitrophenol
2
4,6-Dinitro-o-cresol
3
Phenol
4
4-Nitrophenol
5
p-cresol
6
o-cresol
7
creosol
8
2-Chlorophenol
9
2-Nitrophenol
10
P-Chlorophenol
11
2,4-Dimethylphenol
12
2,4,6-Trichlorophenol
13
Pentachlorophenol
14
2,4-Dichlorophenol
15
2,4,5-T richlorophenol
A = Tyrosinase (3.0%) CPE O -0.20V
B = Plain CPE O + 1.20V
8 min
8nA

s4
^2
6&7 3
I B
Figure 4-2. Chromatograms of 15-component phenol mixture at pH 6.5, under optimum experimental
conditions. Conditions: Mobil* Phaam 25:75 Acetonltrlle: PBS pH 6.5, 0.05M, Flow Rata-. 0.5
ml/mln, Column: BAS C-18, Loop Size: 20/d, Sample: 0.1 mM solution phenol compounds,
Datactor. Dual Electrochemical Detector In parallel (tyrosinase (3.0%) CPE at -0.2V and Plain CPE
at +'1.2V).
8

-------
<
c
s>
k_
3
o
CALIBRATION PLOT
i
10
15
r
20
25
Concentration (|iM)
Figure 4-3. Typical calibration plot* for 2,4-dlnitrophenol and phenol at tyrosinaae (3.0%) carbon paste
electrode, using optimal experimental condltlone.
Application of the new LCEC protocol for the
analysis of spiked environmental samples is shown in
Figure 4-5. Both the groundwater (c) and creek
water (d), from the Hanford and Oak-Ridge
locations, respectively, offer convenient quantitation
of the four spiked phenols (at the 12.5/^M level). No
other peaks are observed at both electrodes,
reflecting the high specifity of the method using
these matrices. Recovery values (at the enzyme
electrode) for these phenols in the groundwater
sample range from 107 to 119%, while in the creek
water sample from 103 to 114%. The peak-current
ratio values (L-A^). using the groundwater and
creek water samples are 0.99 and 0.96 (p-
nitrophenol), 0.78 and 0.78 (2,4-dintrophenol), 0.84
and 0.93 (cxresol) aid 0.87 and 0.79 (phenol), respectively.
9

-------
CALIBRATION PLOT
12-
10-
p-Nitrophenol
o-Cresol
~i
20
Concentration (|iM)
Figure 4-4. Typical calibration plota tor p-nitrophanol and o-craaol at tyrosinase (3.0%) carbon paata
electrode, uaing optimal axparimantal conditions.
10

-------
Peak#
Compound
1
2-dinitrophenol
2
Phenol
3
p-nitrophenol
4
o-cresol
B
a
JH
<
C
00
8 min
I!
Figure 4-5. Sample analyses. Typical chromatograms of real water samples spiked with a 12.5/^M four-component phenol mixture
under optimum experimental conditions. Conditions: Mobile Phaser. 25:75 Acetonitrile: PBS (pH 6.5,0.05M), Flow Rate:
0.5 mVmin, Column: BAS C1t> Loop Size: 20pi, Detector: Dual Electrochemical Detector in parallel (tyrosinase (3.0%)
CPE, -0.2V; Plain CPE, +1-2V), Sample: a: Typical Injection of 12.5/JM phenol mixture onto column, b: 50% Hanford
Ground Water, WS, spiked with 12.5 (M phenol mixture, c: Hanford Ground Water, WS, spiked with 12.5 (M phenol
mixture, d: Oak Ridge Creek Water, TN, spiked with 12.5 /M phenol mixture, I: Injecting sample, A: tyrosinase CPE; B:
Plain CPE.

-------
Chapter 5
Conclusions and Recommendation
Overall, this project has demonstrated that enzyme-
based biosensors for phenols are compatible with the
monitoring of these pollutants in liquid-
chromatographic effluents. The tyrosinase-based
detector enhances the information content of
chromatographic measurements of phenolic
contaminants, and facilitates their monitoring in
relevant environmental samples. Such a detector can
be employed also in a flow-injection system, that
would provide a rapid screening for the "total" phenol
content. Indeed, one can switch between the flow
injection and liquid-chromatographic systems (using
an appropriate valve) to obtain the rapid screening
first, and whenever needed (for elevated levels) switch
to the more detailed chromatographic analysis.
12

-------
References
1.	Achilli, G., Cellerino, G., d'Eril, G., and Bird, S.,	3. Wang, J., Lu, F., and Lopez, D., Analyst,
J. Chromatogr. 697(1995)357.	119(1994)455.
2.	Wang, J., Lu, F., and Lopez, D., Biosesnors &	4. Wang, J. and Lin, Y., Anal. Chim. Acta,
Bioelecton. 9(1994)9.	271(1993)53.
13

-------