Detection of polycyclic aromatic hydrocarbons using quartz crystal microbalances

Article abstract of DOI:10.1021/ac0257546

PAH are a complex class of hydrocarbons resulting from the incomplete combustion of organic substances such as fossil fuels, coal, fuel oil, gas, etc. A chemical coated piezoelectric sensor was developed to determine PAH (e.g., anthracene, naphthalene) in the liquid phase. The PAH derivative 9-anthracene carboxylic acid was used, which once bound to the alkane thiol functions as the recognition element. Binding of anthracene via π-π interaction was observed as a frequency shift in the quartz crystal microbalance with a detectability of the target analyte of 2 ppb and a response range of 0-50 ppb. Such a sensor could be used to identify key marker compounds and give an indication of total PAH flux in the environment. The mass sensitivity of the acoustic device could be improved by increasing the resonance frequency of the sensor because the sensor response to mass loading is proportional to the square of the operating frequency.

Full text of DOI:10.1021/ac0257546

Anal. Chem. 2003, 75, 1573-1577
De t e c t io n o f P o lyc yc lic Aro m a t ic Hyd ro c a rb o n s
Us in g Qu a rt z Crys t a l Mic ro b a la n c e s
S. Sta nle y,† C. J . Pe rc iva l,*,† M. Aue r,† A. Bra ithw a ite ,† M. I. Ne w ton,† G. Mc Ha le ,† a nd W. Ha ye s ‡
Department of Chemistry and Physics, The Nottingham Trent University, Nottingham NG11 8NS, U.K., and
Department of Chemistry, University of Reading, Whiteknights, Reading RG6 6AD, U.K.
electron systems of these aromatic compounds determines their
physical and chemical properties. Naphthalene, for instance, a two-
ringed compound, is the smallest member of the class and is found
in the vapor phase in the atmosphere, while PAHs consisting of
three to five rings are present in the air in both the vapor and
particulate phases. The PAHs with five or more rings are
commonly found as solids adsorbed onto particulate matter in the
atmosphere.11
The flat hydrophobic shape and high lipid solubility of PAHs
makes their excretion from the body difficult. The PAH is then
able to intercalate with the structure of DNA and, consequently,
disrupt the proper functioning of DNA.12 More often than not,
such damage is repaired; however, if the mistake is not re-
paired, the DNA can be misread, resulting in defective pro-
teins, which may ultimately lead to cancer.12 The possible health
hazards associated with PAH exposure are numerous and include
skin problems, immunodeficiency, reproductive difficulties and
cancer.13
A chemically coated piezoelectric sensor has been devel-
oped for the determination of P AHs in the liquid phase.
An organic monolayer attached to the surface of a gold
electrode of a quartz crystal microbalance (QCM) via a
covalent thiol-gold link complete with an ionically bound
recognition element has been produced. This study has
employed the P AH derivative 9-anthracene carboxylic acid
which, once bound to the alkane thiol, functions as the
recognition element. Binding of anthracene via π-π
interaction has been observed as a frequency shift in the
QCM with a detectability of the target analyte of 2 ppb and
a response range of 0 -5 0 ppb. The relative response of
the sensor altered for different P AHs despite π-π interac-
tion being the sole communication between recognition
element and analyte. It is envisaged that such a sensor
could be employed in the identification of key marker
compounds and, as such, give an indication of total P AH
flux in the environment.
Although the effects of many PAHs on mammals is poorly
understood, a number are known or suspected carcinogens. The
five-ringed PAHs benzo[a]pyrene and dibenz[a]anthracene are
known for their carcinogenicity. Benzo[a]pyrene itself is regarded
as a key indicator compound of PAHs because it occurs in all
mixtures where PAHs are present. The U.S. EPA has established
maximum contaminant levels (MCLs) for public drinking water
to minimize the risk of adverse health effects. The MCL for total
PAHs is 0.2 ppb. In light of the obvious health effects associated
with PAHs and their ever increasing use and emission, there is a
clear need to monitor these compounds.
Polycyclic aromatic hydrocarbons (PAHs) are a complex class
of hydrocarbons resulting from the incomplete combustion of
organic substances, such as fossil fuels1, coal,2,3 fuel oil,4 gas,5
refuse,6 wood,7 tobacco,8 and foodstuffs.9 Additional sources
include wood preservative plants, incinerators, asphalt road and
roofing operations, and aluminum plants.10 The widespread
generation and subsequent release of PAHs into the environment
means that they are found in air, water, and soil throughout the
world, often as a complex mixture that may exist for months or
years.5
Our increasing impact on the environment demands that we
establish faster, cheaper and more accurate ways to analyze a wide
variety of compounds. Gravimetric sensors, such as the quartz
crystal microbalance (QCM), are well suited as transducer
elements for chemical sensors, being portable, rapid, and sensitive.
For applications in chemical sensing, a recognition element is
added to the acoustic wave device capable of selectively binding
the analyte to the device surface. The response of these devices
PAHs are characterized by their fused-ring structures consist-
ing of between two and seven aromatic rings. The conjugated π
* Corresponding author. E-mail: carl.percival@ntu.ac.uk.
† The Nottingham Trent University.
‡ University of Reading.
(1) Wayne, R. P. Chemistry of the Atmospheres, 3rd ed.; Oxford University
Press: Oxford, 2000.
(2) Vo-Dinh, T.; Martinez, P. R. Anal. Chim. Acta 1 9 8 1 , 125, 13-19.
(3) Mumford, J. L.; Helmes, C. T.; Lee, X.; Seidenberg, J.; Nesnow, S.
Carcinogenesis 1 9 9 0 , 11, 397-403.
(4) Benner, B. A., Jr.; Gordon, G. E.; Wise, S. A. Environ. Sci. Technol. 1 9 8 9 ,
(9) Crosby, N. T.; Hunt, D. C.; Philip, L. A.; Patel, I. Analyst 1 9 8 1 , 106, 135-
145.
23, 1269-1278.
(5) Finlayson-Pitts, B. J.; Pitts, J. M. Chemistry of the Upper and Lower Atmosphere,
1st ed.; Academic Press: New York, 1999.
(10) Vo-Dinh, T. Chemical Analysis of Poly Aromatic Hydrocarbons, 1st ed.; John
Wiley and Sons: New York, 1989.
(6) Lee, W. M. G.; Yuan, Y. S.; Chen, J. C. J. Environ. Sci. Health 1 9 9 3 , A28,
1017-1035.
(7) Alfheim, I.; Ramdahl, T. Environ. Mutagen. 1 9 8 4 , 6, 121-130.
(8) Gundel, L. A.; Mahanama, K. R. R.; Daisey, J. M. Environ. Sci. Technol.
1 9 9 5 , 29, 1607-1614.
(11) Junge, C. E. Adv. Environ. Sci. Technol. 1 9 7 7 , 8, 7-26.
(12) Hoffman, D. J.; Rattner, B. A.; Burton, G. A. Jr.; Cairns, J. Jr. Handbook of
Ecotoxicology; Lewis Publishers: London, 1995.
(13) Smith, L. E.; Denissenko, M. F.; Bennet, W. P.; Li, H.; Amin, S.; Tang, M.;
Pfeifer, G. P. J. Natl. Cancer Inst. 2 0 0 0 , 92, 803-811.
10.1021/ac0257546 CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/25/2003
Analytical Chemistry, Vol. 75, No. 7, April 1, 2003 1573

S c h e m e 1 . Ou t lin e o f t h e S yn t h e s is o f t h e Io n ic
Co a t in g
Fig u re 1 . Schematic diagram of the π-π-stacking interaction in
coating on crystal surface.
is then based on a decrease in their resonance frequency as mass
is attached to the device or the recognition element.
Recently, Dickert et al.14 showed that a molecularly imprinted
polymer-coated QCM displays affinity toward a range of PAHs.
In this work, an alternative approach was used that relies upon
the formation of an alkanethiol monolayer upon the surface of
the transducer to which is attached the recognition element. It is
possible to form a mixed monolayer of two alkanethiols with
different tail groups when depositing from a solution onto a surface
rather than phase segregating into macroscopic domains.
To achieve a specific adsorption of PAHs onto the organic
monolayer, PAH acid analogues were attached to the other end
of alkanethiol, as indicated in Figure 1. The appended PAH groups
form a π-π stacking array. Other PAHs from a surrounding PAH
solution bind to the PAH thiol layer via π-π interaction between
the π-electron systems of aromatic multirings. This electronic
interaction, characteristic of aromatic systems, can occur when
the π bonds of an aromatic molecule undergo spontaneous
polarization. The spontaneous dipole that results induces a second
dipole on another aromatic molecule and they are attracted to
each other,15 with the additional mass bound in the organic layer
detected by the QCM. This interaction is likely to occur with any
PAH in the solution, but most favorably with PAH-molecules that
have a structure that is identical to the aromatic system employed
as recognition element. Thus, the necessary selectivity is achieved
for this coating to be effective in terms of its ability to interact
with the target analyte via a π-π stacking array.
plano-plano quartz crystals with Cr/ Au contacts, operating at a
fundamental resonance frequency of 5 MHz and with an electrode
area of ∼133 mm2 (Maxtek model no. 149211-2). The crystals were
mounted in a Maxtek CHC-100 crystal holder, and the measure-
ment system consisted of a Maxtek PLO10 phased-locked oscil-
lator and an Agilent HP53132A universal counter interfaced to a
microcomputer.
In this work, a piezoelectric sensor with a preferential selectiv-
ity for the PAH anthracene is presented. An organic monolayer
containing the recognition element is attached to the surface of a
gold electrode of a quartz crystal microbalance (QCM) via a
covalent thiol-gold bond.
EXPERIMENTAL SECTION
Prior to the application of the polymer coating to the electrode
area, the crystals were prepared using pirannha etch solution (1:3
30% v/ v H2O2/ conc H2SO4). The crystals were immersed in
Pirannha etch solution for 10 minutes, then rinsed with deionized
water and ethanol and dried overnight.
Reagents. 3-Dimethylaminopropyl chloride hydrochloride,
potassium carbonate, anthracene, potassium thioacetate, 1-bu-
tanethiol, 9-anthracene carboxylic acid, naphthalene, benzo[a]-
pyrene and 1,2:5,6-dibenzanthracene were purchased from Aldrich
Chemicals (Dorset, U.K.). All the solvents were of analytical grade
and all reagents were used as supplied.
Ionic P AH Coating. The synthesis of the ionic coating is
summarized in Scheme 1. A 0.01-mol portion of 3-dimethylami-
nopropyl chloride hydrochloride (CH3)2N(CH2)3Cl‚HCl was stirred
together with 0.01 mol potassium carbonate (K2CO3) in 20 mL of
dried acetone under nitrogen for ∼0.5 h. Subsequently, 0.01 mol
of potassium thioacetate (KSCOCH3) was added, stirred for ∼2.5
h, and left overnight to form a precipitate. The solution was filtered
Quartz Crystal Microbalance. The quartz crystal microbal-
ance was produced using unpolished AT-cut, 25-mm-diameter,
(14) Dickert, F. L.; Besenbo¨ ck, M.; Tortschanoff, M. Adv. Mater. 1 9 9 8 , 10, 149.
(15) Israelachvili, J. Intermolecular and Surface Forces, 2nd ed.; Academic
Press: New York, 1992.
1574 Analytical Chemistry, Vol. 75, No. 7, April 1, 2003

off and evaporated, yielding a precipitate as the product of step 1,
which was identified by NMR spectroscopy.
The acetate group was replaced by a sodium ion in step 2 by
dissolving 0.3 g (∼2 mmol) of diethylaminopropylthioactetate in
15 mL of 4 M sodium hydroxide-methanol solution. This solution
was mixed with 30 mL of 2 M HCl and half the molar amount of
1-butanethiol in a glass beaker containing a quartz crystal. The
acid neutralizes the hydroxide, and the thiolate ion is protonized,
forming a thiol group (step 3) that subsequently forms a covalent
bond with the gold of the electrode on the crystal surface (step
4). Simultaneously, the butanethiol fixes on the gold electrode in
the same manner, forming spacers between the aminothiols. The
crystals were left in this solution for ∼4 h, then they were taken
out of the solution. To ionically attach an anthracene molecule,
the quartz crystals were placed directly into a solution of the same
molar amount of 9-anthracenecarboxylic acid as aminothiol in ∼30
mL of ethanol for ∼1 h. Anthracenecarboxylate ions are assumed
to form an ionic link with quarternary ammonium ions on top of
the organic chains on the crystal (step 5). The crystals were rinsed
with ethanol after having been taken out of the solution, then dried
with and kept under nitrogen until they were screened.
Fig u re 2 . Resonance frequency change of the QCM with the ionic
coating: coating 1 ()), coating 2 (0), coating 3 (4), coating 4 (O),
coating 5 ((), and coating 6 (9).
Reference crystals were made by covalently binding the
mixture of butanethiol and aminothiols to the quartz crystals. This
is achieved by following the outlined synthesis and finishing at
step 4. At this point, control samples (blank crystals) were washed
with ethanol and dried under nitrogen.
Evaluation of Sensor Response. Solutions containing known
amounts of the analyte were prepared in ethanol. The coated
quartz crystals were placed into a beaker of ethanol until a stable
response was obtained, then 9 mL of analyte solution was added
to the stirred bulk ethanol solution via successive 1 mL aliquots.
The frequency was recorded until a stable response was obtained.
After each 9-mL addition of analyte, the coated quartz crystal was
soaked in the porogenic solvent for 10 min, and the experiment
was repeated.
A drift in resonance frequency of the QCM may be observed,
which is caused by changes in the environmental conditions, such
as temperature. On the time scale of the experiment, the change
in frequency caused by environmental conditions could be greater
than that caused by the mass loading of the quartz crystal. To
overcome this difficulty, the frequencies of two quartz crystals
immersed in the ethanol solution, one crystal with the ionic coating
and one crystal with no attached PAH group coating were
measured simultaneously following the DQCM technique of
Bruckenstein et al.16 The resultant difference in frequency between
the two crystals can be used to assess the response of the crystal
to mass loading of the analyte, because the crystal with no PAH
coating displayed no detectable change in resonance frequency
with mass loading; all experiments were performed in this manner.
Fig u re 3 . Resonance frequency change of the QCM with the ionic
coating: run 1 ()), run 2 (0), run 3 (4), and after replenishing coating
(O).
gradient in the range Santhracene ) -2.047 ( 0.042 Hz ppb-1
,
which reaches saturation at concentrations exceeding 20 ppb.
Similar sorption isotherms have been observed for the binding
of analytes to a polymer with a finite number of interaction sites.17
The data presented in Figure 2 shows the response for six
different quartz crystals, each with a PAH coating produced in
the same way. The agreement suggests that our methodology for
coating the quartz crystals is reproducible between different
coatings.
Between screenings, the crystal was thoroughly rinsed with
ethanol. This cleaning method, however, seems to have been only
partly successful. Subsequent test runs with one crystal showed
in general a greatly reduced frequency shift, as shown in Figure
3. This implies that the number of analyte interaction sites
therefore has been reduced, many of them remaining blocked
after the initial screening.
The ionic link that holds the aromatic groups in place can be
broken, and another analyte interaction site can be attached. To
examine this, coated crystals that had already been screened were
placed first into a NaOH solution, then into a 9-anthracenecar-
boxylic acid-ethanol solution. Applying 4 M NaOH for ∼1 h and
9-anthracenecarboxylic acid overnight was performed in order to
replenish the coating. Figure 3 also shows a comparison between
the fresh coating and the replenished coating. It can be seen that
RESULTS AND DISCUSSION
Response of the Sensors to Mass Loading of Anthracene.
The change in resonance frequency of the coated quartz crystal
as a function of anthracene concentration is shown in Figure 2.
There is an initial linear decrease in frequency corresponding to
an increase in anthracene concentration up to ∼15 ppb, with a
(16) Bruckenstein, S.; Michalski, M.; Fensore, A.; Li, Z.; Hillman, A. R. Anal.
Chem. 1 9 9 4 , 66, 1847-1852.
(17) Percival, C. J.; Stanley, S.; Galle, M.; Braithwaite, A.; Newton, M. I.; McHale,
G.; Hayes, W. Anal. Chem. 2 0 0 1 , 73, 4225-4228.
Analytical Chemistry, Vol. 75, No. 7, April 1, 2003 1575

Ta b le 1 . S u m m a ry o f t h e Ex p e rim e n t a lly Eva lu a t e d Ap p a re n t Dis s o c ia t io n Co n s t a n t s a n d In it ia l S lo p e s
constant
anthracene
naphthalene
benzo[a]pyrene
1,2:5,6-dibenzanthracene
S
initial/ ppb-1
-2.047 ( 0.042
60.16
-0.988 ( 0.048
51.94
0.327 ( 0.019
47.22
0.437 ( 0.024
31.26
K, ppm-1
Naphthalene has a smaller value of apparent dissociation
constant than anthracene. This implies that either naphthalene
has a lower molecular mass than anthracene and that the same
number of bound naphthalene molecules, therefore, results in a
smaller additional mass in the coating and, thus, a smaller
frequency shift or that the anthracene groups of the coating show
a lower affinity to the somewhat different π-electron system of
the naphthalene molecules. The difference in the size of the
frequency shift between runs with anthracene and naphthalene
gives no definite indication which effect is predominant. However,
runs with benzo[a]pyrene and 1,2:5,6-dibenzanthracene show
similar results. Both are five-membered PAHs and, therefore, have
a higher molecular weight than anthracene but still show a
decidedly lower response, as indicated by the low values of the
apparent dissociation constant. This suggests that the coating
selectively binds the target molecule, anthracence, over the larger
PAHs, benzo[a]pyrene or 1,2:5,6-dibenzanthracene. The coating
has less affinity for the larger PAHs, and this effect overrides the
influence of their comparatively larger molecular weight. This
indicates that the coating displays a degree of shape selectivity.
Dickert et al.14 produced recognition element for PAHs via the
formation of a molecular imprinted polymer of anthracene-coated
QCMs. Their results indicated that the MIP recognition element
displayed an affinity to several PAHs and showed a greater affinity
for benzo[a]pyrene, 1,2:5,6-dibenzanthracene, chrysene, pyrene,
and benzopyrelene over the target molecule anthracene, with the
greatest affinity to chrysene. Unfortunately, it is not possible to
directly compare our values of the apparent dissociation constant
with those obtained by the molecular imprint approach of Dickert
et al., because the values are not given. However, it can be seen
that both the MIP approach and our ionic coating display an
affinity to PAHs and that both approaches display some selectivity
in terms of size and shape of PAH.
Fig u re 4 . Range of monoterpene analogues that the selectivity of
the ionic coated quartz crystals were tested toward: anthracene ()),
naphthalene (0), 1,2:5,6-dibenzanthracene (4), and benzo[a]pyrene
(O).
the measured frequency shift was still clearly below its original
value. Therefore, it can be seen that under the experimental
conditions used, it is not possible to recover 100% of the analyte
interaction sites. All subsequent analysis is performed on fresh
coatings.
Selectivity. To examine the selectivity of the created sensor,
crystals with the same coating with anthracene as a functional
group were screened with different PAHs, as shown in Figure 4.
All curves show linear behavior until the concentration reaches
∼15-20 ppb, which reaches saturation at concentrations exceed-
ing 20 ppb, the gradients of which are given in Table 1.
The affinity of the PAH ionic coatings toward PAHs can be
evaluated using a one-site Langmuir-type binding isotherm anayl-
sis,18
CONCLUSIONS
∆f
∆f
It is clear that quartz crystal sensors coated with a monolayer
Kc
+
) Kc
∆f∞ ∆f∞
of aminothiols with an ionically bound anthracene group can be
used to preferentially detect the PAH anthracene in the liquid
phase with a sensitivity of 2 ppb. There are several possibilities
to further improve the sensitivity.
First, it may be possible to increase the number of interaction
sites on the QCM. To achieve this, the synthesis of the coating
may be improved to give a higher yield. For instance, the effect
of different amounts of the spacer 1-butanethiol may also be
examined in order to achieve an optimal accessibility of the
interaction sites.
Second, the mass sensitivity of the acoustic device can be
improved. This may be achieved by increasing the resonance
frequency of the sensor, because the sensor response to mass
loading is proportional to the square of the operating frequency.
The upper limit for quartz crystal microbalances operating at the
fundamental frequency is around 10 MHz; the higher the
where K is the apparent dissociation constant, c is the free analyte
concentration at equilibrium, ∆f is the frequency change, and ∆f∞
is the frequency corresponding to complete coverage. Although
this does not reflect the real situation in the polymer, it allows
for a comparison of binding between polymers. Using the data
shown in Figure 4, the apparent dissociation constants K were
estimated and are given in Table 1.
Anthracene shows the largest apparent dissociation constant,
which indicates that the anthracene groups of the coating attract
the anthracene molecules in the solution well. The reason for this
behavior is assumed to be the nearly identical π-electron systems
of the dissolved molecules and the coating.
(18) Yilmaz, E.; Mosbach, K.; Haupt, K. Anal. Commun. 1 9 9 9 , 36, 167-170.
1576 Analytical Chemistry, Vol. 75, No. 7, April 1, 2003

frequency, the thinner the crystal. Alternatively, surface acoustic
wave devices may be employed that will extend the range of
operating frequencies to over 1 GHz.19,20
For applications of the coated quartz crystal sensors in the
field, future work to be considered would be to look for a way to
break the ionic link that holds the PAH group and to replenish
the coating with the same or a different group, creating a sensor
that is capable of detecting different PAHs, such as benzo[a]-
pyrene, which is considered to be the key marker compound for
PAHs. The sensor is likely to respond to other environmental
analytes; however, the ability of the sensor to respond preferen-
tially to the target analyte compared with closely related analogues
offers the potential to distinguish between PAHs and other
analytes.
(19) Ji, H. S.; McNiven, S.; Ikebukuro, K.; Karube, I. Anal. Chim. Acta 1 9 9 9 ,
Received for review May 7, 2002. Accepted January 29,
2003.
390, 93-100.
(20) Jakoby, B.; Ismail, G. M.; Byfield, M. P.; Vellekoop, M. J. Sens. Actuators, A
1 9 9 9 , 76, 93-97.
AC0257546
Analytical Chemistry, Vol. 75, No. 7, April 1, 2003 1577