Tải bản đầy đủ - 0 (trang)
2…Optical Properties and Their Exploitation in Sensing

2…Optical Properties and Their Exploitation in Sensing

Tải bản đầy đủ - 0trang


Optical Sensors and Probes


change in the optical response is reagent-mediated [4]. Commonly used direct and

reagent-mediated optical sensing approaches for the most important types of

analyte are summarised in Table 12.1. Given that direct sensing generally falls

within the scope of direct spectroscopic analysis, here we will focus our discussion

on reagent-mediated sensors. Since these do not require the analyte to exhibit a

measureable optical parameter, they are particularly useful for those analytes, or

concentrations, where direct sensing will not work. Reagent-mediated sensing is

therefore suitable for a much wider range of analytes than direct sensing.

The response of an optical sensor or probe is determined from the change in its

optical properties in the presence/absence of the analyte. When light strikes a

solution or a solid containing a molecular probe, a number of different interactions

are possible. If absorbed in an electronic transition, then the resultant excited-state

will then be available to react with other species or to relax radiatively or nonradiatively. If the probe is immobilised within a solid support matrix, then scattering, refraction and reflection of the incident light may be also be important.

Alternatively, the primary response signal may arise from these light interactions

at a responsive surface, rather than from a combination of interactions at discrete

probe sites. The underlying principle of optical sensing is that the presence of the

analyte affects the rate and/or efficiency of one or more of these processes to some

extent. Since light has a number of measureable properties, such as wavelength,

intensity, phase and polarisation, which are easily monitored with spectroscopic

techniques, it is possible to correlate the change in any of these optical parameters

with the analyte concentration. Absorption and photoluminescence are most

commonly used to monitor the optical response and we will now consider these in

more detail.

12.2.1 Absorption

The presence of the analyte may cause a change (e.g. a spectral shift, intensity

change, formation of new band) in the absorption spectrum of the probe. This may

be monitored using a spectrometer to quantitatively determine the analyte concentration. The detection limit will be determined by the probe sensitivity and the

instrument limitations but *10-5–10–7 mol dm-3 is typical. When the absorbance

change is a wavelength shift in the visible region it may cause an observable

colour change, enabling qualitative or even semi-quantitative analyte detection by

sight alone—i.e. a colorimetric sensor. This offers a distinct advantage in situations where rapid assessment of the analyte presence is required. The most familiar

colorimetric probe is perhaps the pH universal indicator strip, which is impregnated with halochromic dyes, whose absorption properties, and therefore colour,

are modified by pH.







Complexation PET/






fluorescence Luminescence Conformational shifts



Complexation Heavy

Luminescence atom quenching PET/




Luminescence Complexation


Complexation PET/

Luminescence RET Heavy atom


PET photoinduced electron transfer, RET resonance energy transfer, ICT intramolecular charge transfer

DNA, proteins


Cl-, Br-, I-, NO2-,

NO32-, H2PO4-

Glucose, fructose



Na+, Ca2+, Mg2+

Fe3+, Cu2+, Zn2+ Pb2+,

Hg2+, Cd2+

Petroleum oils

Thiols, phenols





NO2, H2S, HCN, NO,

anaesthetic gases







CO2, NH3, SO2

Small molecule gas (pKa

active and pKa used in


Small molecule gas (non-pKa

active, or pKa not used in


Metal cations

Alkali/alkaline earth

Transition metal

Heavy metal



Complexation Heavy

Luminescence atom quenching



pKa equilibrium

Luminescence Luminescence [H+ measured]


Elemental gas



Luminescence Quenching

Spectral shifts


Spectral shifts



Spectral shifts



Spectral shifts


Spectral shifts



Spectral shifts


Spectral shifts


Spectral shifts















Blood gas

Blood gas




Blood gas


Table 12.1 Examples of direct and reagent-mediated optical sensing methods for important analyte types. Note that not all direct or reagent-mediated

methods indicated are equally useful for each example of analyte type

Analyte type



ReagentIntermediate stage

Sensing method




R. C. Evans and P. Douglas


Optical Sensors and Probes


12.2.2 Photoluminescence

Photoluminescence can be used to detect an analyte in three ways: (1) the analyte

itself is intrinsically fluorescent (direct sensing); (2) the analyte can be tagged with

a fluorophore label; or (3) the analyte interacts with a luminescent probe. Direct

sensing and fluorophore-tags are widely used in biomedical applications to probe

cell environments. Many proteins are intrinsic fluorophores due to the presence of

the aromatic amino acids tryptophan, phenylalanine and tyrosine. Analytes such as

pH, CO2, NH3, O2 and various cations and anions can be measured indirectly using

luminescence probes.

Photoluminescence measurements are inherently more sensitive than absorption, enabling detection limits of ~10–9 mol dm-3 to be readily achieved. Luminescence intensity and lifetime are the most commonly monitored properties;

however fluorescence anisotropy, spectral shifts, and changes in vibrational finestructure may all be used as probing parameters.

Intensity probes may be either fluorescent or phosphorescent. Often, the analyte

causes a decrease, or quenching, of the emission intensity, which, for simple

systems may be related to the analyte concentration using the Stern–Volmer

equation (see Chap. 1). However, deviation from linear Stern-Volmer kinetics is

common and more complex models are often required to calibrate the probe

response. Simple intensity measurements may require only an excitation source

and detector. However, these measurements are susceptible to signal drift due to

fluctuations in the source intensity and variations in the probe concentration due to

photodegradation or leaching. This problem can be overcome using wavelengthratiometric detection, where an analyte-insensitive excitation or emission band is

used as a reference. Spectral changes, such as wavelength shifts and variation of

vibrational structure may provide information on the nature of the probe-analyte

interaction. For example, the emergence of vibronic structure in the emission

spectrum may indicate the formation of a probe-analyte complex, whereby the

probe adopts a more rigid or ordered conformation in the presence of the analyte.

Similarly, shifts in the emission wavelength may be used to identify changes in the

local environment, such as polarity, of the probe.

Lifetime measurements offer the advantage of being unaffected by fluctuations

in the excitation source intensity or probe concentration. Since triplet excited states

typically have much longer radiative lifetimes than singlet states, they are particularly susceptible to interactions with other molecules. Quenching interactions

can also be investigated using lifetime measurements and modelled using the s0/s

form of the Stern-Volmer equation. The form of the Stern-Volmer plot may also

indicate whether the probe-quencher interaction occurs predominantly via static or

dynamic quenching (see Chap. 1).

Fluorescence anisotropy may also be used as a response parameter. Upon

excitation with polarised light, many fluorophores give polarised emission, with

the extent of polarisation being given by the anisotropy, r. Changes in anisotropy,

as a consequence of changes in probe tumbling, e.g. by probe binding or


R. C. Evans and P. Douglas

unbinding, or changes in emission lifetime when emission and tumbling rates are

comparable, can be used as probe response. Since polarisation anisotropy requires

emission measurements at both vertical and horizontal polarisations the method

already has a ratiometric character, and the long-term stability of the excitation

source is less important than for analysis using a single intensity measurement. A

ratiometric approach involving two lumophores has also been proposed, primarily

because of the wide dynamic range it offers. In this method the analyte-sensitive

probe, P, and the analyte-insensitive reference, R, are both lumophores, but exhibit

different anisotropy values when isolated. The combined fluorescence from both

species is used to determine the analyte concentration [6]. The measured anisotropy of the steady-state emission of a mixture of lumophores is given by the

intensity-weighted average of the individual anisotropies, rP and rR:

r ẳ rP fP ỵ rR fR


where fP and fR are the fractional intensities of the two fluorophores and

fP ? fR = 1. Since the fluorescence intensity and anisotropy of R are constant, any

change in either fP or rP will result in a change in the measured anisotropy. To

maximise the range of anisotropy values attainable, the reference fluorophore is

chosen to have a complementary anisotropy of either *0 or *1, depending on the

relative anisotropy of the probe, such that in combination, a dynamic range of

almost one full unit (i.e. 0–1) is possible.

12.3 Probe Response Mechanisms

The precise response mechanism will depend on the probe-analyte combination

and is often difficult to assign unambiguously. However, the response of most

probes arises from one (or more!) of the following:

• analyte-induced change to the probe molecular structure;

• introduction of a competing kinetic process;

• physical change in the local environment of the probe.

12.3.1 Analyte-Induced Change to the Probe Molecular


The most obvious mechanism by which the analyte may modify the molecular

structure of the probe is the formation of an analyte-probe complex. The analyte

may bind to the probe by either physical or chemical bonding. For reversible

sensing the binding interaction should be completely reversible, otherwise the

probe will be suitable only for single-shot measurements. Knowledge of the

binding equilibria is critical when designing sensors based on this type of probe.


Optical Sensors and Probes

409 Binding Equilibria Theory

The theoretical basis for analyte detection via binding is the same regardless of the

actual binding mechanism. If we consider a sensing medium, where the relative

concentrations of the analyte, A, and the probe, P, are such that the analyte is

present in excess so that binding does not alter analyte concentration, then an

equilibrium will exist between the bound (PB) and free (PF) states of the probe. If

the binding stoichiometry is 1 to 1, the dissociation equilibrium is given by:

P B ) A ỵ PF


and the dissociation constant, Kd, is defined as,

Kd ẳ





The fraction of bound states of the probe is given by:




Kd ỵ ẵA


where [P] is the total concentration of the probe (i.e. [P] = [PF ? PB]). An

important result arises from this relationship: if a probe is to give a useful response

the binding constant of the probe must be comparable to the analyte concentration.

The useful range of analyte concentrations is typically restricted to 0.1Kd \ [A] \

10 Kd [1]. Analyte concentrations outside of this range will produce only a limited

change to the measured signal.

Where the instrumental response is the sum of two terms, one each for PF and

PB, and each of these terms is linearly dependent upon the respective concentration

then, irrespective of the nature of the response, the analyte concentration may be

obtained from:

S À Smin

½AŠ ¼ Kd


Smax À S

where Smin is the instrument signal when all the probe is free (zero analyte), Smax is

the instrument signal when all the probe is complexed (suitably high analyte

concentration), and S is the instrument signal at the analyte concentration of

interest, i.e. when the probe is partially-bound. If response is not the sum of two

terms, or if these terms are not linearly dependent upon the respective concentrations, then Eq. 12.5 does not apply and a more complex calibration equation is


Changes in absorption or emission intensity

Changes in absorbance can be used with Eq. 12.5, provided absorption by both

bound and unbound probe obey the Beer-Lambert law across the wavelength band

of measurement.


R. C. Evans and P. Douglas

Changes in emission intensity can also be used, but this requires more careful

consideration of the experimental procedure. All measurements must be performed

using the same instrumental configuration (slit widths, excitation wavelength etc.),

optical pathlength and probe configuration. To compensate for variations in

excitation intensity, or geometry, an additional unquenched lumophore which

emits in a different spectral region to the probe can be added to provide an

intensity-ratiometric measurement. Many researchers perform solution phase

measurements using a titration technique, where aliquots of an analyte solution are

sequentially added to a solution of the probe. When using this approach, it is

important to consider and compensate for a reduction in probe concentration, due

to dilution effects.

Spectral shifts

For an optical probe that displays a wavelength shift in absorption or emission

on binding, the analyte concentration may be determined from a ratio of intensities

at two wavelengths. This is known as a wavelength-ratiometric approach and

avoids some of the limitations mentioned above for intensity-based sensing, since,

the measurements are independent of probe concentration. Absorption-, excitationand emission-based ratiometric measurements are all possible.

Chemical and physical binding mechanisms

Binding probes are widely used for sensing in physiological media. Common

analytes include anions, cations and biomolecules such as polysaccharides, DNA

and proteins. The probe must contain a chemical moiety which acts as a receptor

binding site for the analyte and the selectivity will depend on its binding constant

at this site. Chemical binding results in the formation of a formal covalent or ionic

bond between the analyte and the probe, yielding a new chemical species whose

optical properties differ from that of the probe [7]. Physical binding arises due to a

weaker intermolecular interaction between the analyte and the probe, in the form

of van der Waals, electrostatic or hydrogen bonds. Chemical Binding Probes

pH indicators: Colorimetric pH indicators are typically conjugated organic

chromophores which contain a functional group that is pH sensitive due to its

ability to participate in protonation-deprotonation equilibria. The ionised (Ind–)

and unionised (HInd) forms of the indicator are present in a concentration ratio

determined by a logarithmic form of Eq. 12.5, i.e. the Henderson-Hasselbalch


pH ẳ pKa ỵ log




where pKa = log10Ka and Ka is the acid dissociation constant. The two forms of

the dye have different absorption spectra, so the relative concentration of either


Optical Sensors and Probes


form can be measured optically and related to changes in the pH. The dye pKa

indicates the centre of the measureable pH range at which the visible colour

transition occurs. Although the pH response range for an individual indicator is

quite narrow, a wide variety of indicators are available with responses spanning

the entire pH scale [8] and mixed indicators can be used to give a solution or

impregnated paper with a colorimetric response over a very wide pH range. Since

the absorption bands arise due to n–p* and p–p* transitions, the observed colour

change is determined by the degree of conjugation in the ionised and non-ionised

dye forms. Luminescent pH indicators, where ionised and unionised forms of the

indicator have different emission characteristics, are also available.

Carbon dioxide and ammonia probes: Optical sensors for CO2 and NH3 are

based on pH-sensitive colorimetric or luminescent probes. When CO2 dissolves in

water it behaves as a weak acid, due to the following series of equilibria:

CO2gị ) CO2aq:ị


CO2aq:ị ỵ H2 O ) H2 CO3aq:ị


H2 CO3aq:ị ) Hỵ

aq:ị ỵ HCO3aq:ị




3aq:ị ) Haq:ị ỵ CO3aq:ị :


Thus, the pH of the aqueous solution in equilibrium with CO2 gas is determined

by the partial pressure, pCO2, of carbon dioxide and an optical sensor for CO2 may

be designed by utilising a pH sensitive probe, P, which, on reaction with the

protons generated in these equilibria, gives, HP+. The two forms of the probe must

exhibit different spectral characteristics, such that, a detectable change in the

absorption and/or emission profiles is observed upon protonation. Similar arguments may be applied to the development of optical sensors for ammonia based on

the equilibrium:


4aq:ị ) NH3aq:ị ỵ Haq:ị



Two approaches have been adopted for pCO2 sensing, namely (1) wet sensors and

(2) plastic (solid-state) sensors. A wet sensor consists of a pH-sensitive probe dissolved in aqueous bicarbonate buffer solution, which is separated from the gaseous

or liquid test medium by a gas-permeable membrane [9]. In plastic sensors, a polar

pH-probe is immobilised in a thin polymer film. The probe is usually ion-paired with

a lipophilic base such as a tetra-alkyl ammonium hydroxide. This ion-pair combination facilitates compatibility between the two components, whilst simultaneously

eliminating the need for aqueous buffers due to the associated water of hydration.

This makes it possible for these sensors to maintain their sensitivity to CO2, although

they may show some humidity dependence. The requisites of a CO2 probe are a

suitable pKa, a significant change in absorption or emission characteristics upon


R. C. Evans and P. Douglas

protonation, and photostability. The luminescent probe 8-hydroxypyrene-1,3,6-trisulfonic acid (HPTS, Fig. 12.1) (pKa * 7.5), which is highly fluorescent when

unprotonated, satisfies these requirements. State-of-the-art sensors for CO2 based on

HPTS immobilised in a sol–gel matrix can achieve 80 ppm detection limits for

sensing of CO2 gas in the 0.03–30 % range [10].

Saccharide probes: Chemically-binding saccharide probes make use of the

reversible formation of a cyclic ester between a boronic acid group attached to the

probe (Fig. 12.1) and the diol moiety of saccharides. This interaction affects the

Lewis acidity of the boron atom, which can influence the excited-state decay

pathway of the probe. Often, this will result in the activation of a competing

process, such as photoinduced electron transfer or resonance energy transfer (see

Sect. 12.3.2). Spectral shifts in the UV/Vis absorbance spectrum may also be

observed on binding.


Continuous monitoring of blood glucose levels is particularly important for the

management of diabetes. Since tear glucose levels are well-known to be elevated

in hyperglycaemia sufferers, an optical glucose sensor comprised of a boronic acid

containing fluorophore (BAF, Fig 12.1) based on a quarternised quinolinium

nucleus, immobilised in a disposable contact lens has been developed [11]. The

probe was incorporated into the sensor matrix simply by incubating the contact

lens, made from a hydrophillic poly(vinylalcohol) polymer, in a concentrated

aqueous probe solution. The hydrophillicity of the contact lens also enabled ready

diffusion of the aqueous analytes in tears. The method was suitable for continuous

non-invasive monitoring of tear glucose levels in the range 50–1000 mmol dm-3,

which is appropriate for physiological measurements.

Metal cation recognition probes: Some metals, notably the transition metals,

readily form complexes with organic molecules, resulting in changes to optical

properties, most commonly absorbance. This phenomenon is commonly exploited

in standard methods for the spectrophotometric determination of metal cation

concentrations [8] and standard test kits for a variety of metals are commercially

available [12]. More complex probes based on a cation binding site linked to a

fluorophore or chromophore reporter have also been developed [13]. The cation

Fig. 12.1 Structures and characteristics of some typical chemical binding probes


Optical Sensors and Probes


receptor site is often a crown ether (monocyclic polyether) and selectivity to alkali,

alkaline earth and transition metals can be achieved by varying the acceptor size,

denticity or by substitution of the crown oxygen atoms with heteroatoms such as

nitrogen and sulfur. The most stable complexes are usually formed when the

crown diameter closely matches the cation ionic radius.


The development of probes for the toxic heavy metals mercury, cadmium and

lead in biological systems is of special interest. In this context, selectivity is

particularly important since alkali and alkaline earth metals (e.g. Na+, K+, Ca2+,

Mg2+) are usually present at much higher cellular concentrations. A series of

probes based on a fluorescein-like xanthenone reporter and a ‘soft’ thioether

receptor have been reported for the selective monitoring of Hg2+ (Fig. 12.1) [14].

A large fluorescence enhancement is observed on binding to Hg2+ in water, cells

and tissue, due to the closing of an electron transfer deactivation pathway.

Notably, one probe, MG1, which is structurally modified to restrict rotation

between the receptor and emitter units, displays an emission quantum yield of 0.74

in its bound form, affording detection limits down to the 2 ppb range. Physical Binding Probes

Anion recognition probes: The design of anion receptors poses more challenges

than for their cation analogues. Anions have a lower charge to radius ratio than the

isoelectronic cations meaning that electrostatic binding is often less effective.

Anions also adopt a wide range of geometries including spherical (Cl–, F–), linear

(CN–, OH–), trigonal planar (CO32–, NO3–) and tetrahedral (PO43–) [15]. While this

phenomenon can be exploited to design shape-specific receptor sites, it is also

more difficult to distinguish between anions with the same shape. Synthetic anion

receptors may be positively charged or neutral. The positively charged acceptors

are usually protonated nitrogen atoms or metal cations and anion binding occurs

via electrostatic interactions. Neutral receptors bind anions via hydrogen bonding,

ion–dipole interactions or coordinate anions at the Lewis acidic centre of a neutral

organometallic ligand. The directionality of hydrogen bonds makes it possible to

design receptors with specific shapes capable of differentiating between anions

with different geometries. Commonly used hydrogen-bonding receptors include

ureas, thioureas, pyrroles, indoles and triazoles. The addition of an optical reporter

group to the receptor turns it into an optical probe for the anion. Various chromophores and lumophores have been used including anthracene, naphthalene, and

1,8-napthalimide [15]. Dramatic changes in the ground-state absorption properties

of the probe are often observed on anion binding, making them suitable for colorimetric detection [16, 17].

Indicator displacement assays (IDAs): These are a complementary approach to

analyte recognition probes [17]. In the IDA method a colorimetric or fluorescent

indicator is first allowed to bind reversibly to the receptor. A competitive analyte is


R. C. Evans and P. Douglas

then introduced to the system causing the displacement of the indicator from the

host, which in turn modulates the optical response of the indicator.

DNA probes and stains: Fluorescent probes and markers are often used to visualise

and quantify nucleic acids such as DNA and RNA. Commonly used markers include

DAPI (40 -6-diamidino-2-phenylindole), YOYO (homodimer of oxazole yellow

YO), EtBr (ethidium bromide) and AO (acridine orange) (Fig. 12.2) [18]. These

markers are weakly emissive in solution but exhibit a dramatic fluorescence

enhancement on binding to DNA. The origin of this fluorescence enhancement is

hotly debated and may differ for each dye. Two different binding modes are available, which enables site-specific probing of the DNA structure. YOYO, TOTO and

EtBr are all examples of DNA intercalators – their planar fused aromatic ring

structures enable them to slot in between the DNA base pairs, leading to significant

p-electron overlap. DAPI, on the other hand, preferentially binds to AT-rich

sequences in the DNA minor groove. Hydrophobic interactions and/or hydrogenbonding stabilise this binding process. Acridine orange may be used to probe the

secondary DNA structure. It intercalates with double-stranded (ds) DNA as a

monomer, emitting green fluorescence (530 nm), but binds to single-stranded (ss)

DNA as an aggregate which emits orange fluorescence (640 nm).

12.3.2 Introduction of a Competing Process by the Analyte

The second response mechanism is the situation where the presence of the analyte

introduces a competing process which modifies the optical properties of the probe.

Typically, the competing process activates or deactivates an alternative excited-state

relaxation mechanism; consequently this response is limited to emission probes.

Fig. 12.2 Structures and optical characteristics of some common physical binding probes for



Optical Sensors and Probes


Analyte control of competitive relaxation pathways can be induced by a dynamic

collisional interaction between the analyte or analyte-generated quencher, and

the probe emitter, or alternatively by analyte-binding, either directly to the emitter;

at a site adjacent to the emitter; or to an adjacent molecule. It should be noted that

these mechanisms are not mutually exclusive and assignment of the competing

mode responsible for excited-state deactivation is often ambiguous. Discussion of

these mechanisms based on the nature of the competitive process introduced is

convenient. Competition by Perturbation-Induced Intersystem Crossing

Perturbation quenching occurs when orbital overlap between the probe and

quencher allows some property of the quencher to be shared with the probe

excited-state. Neither probe nor quencher is chemically altered in the process. For

example, the quencher may act as a heavy atom, causing non-radiative relaxation

of the excited singlet state, S1, to the T1 state via intersystem crossing and/or

enhanced radiative relaxation from T1. Detection may then be based on quenching

of S1, i.e. fluorescence quenching, or phosphorescence from the T1 state.


Perturbation quenching is particular useful for detection of heavy atom halides

and halocarbons [19]. The first optical sensor for halides was based on the fluorescence quenching of acridinium and quinolinium probes immobilised on a glass

surface [19]. The detection limits followed the order I– (0.15 mmol dm-3) \ Br–

(0.40 mmol dm-3 \ Cl– (10.0 mmol dm-3). Two possible quenching mechanisms have been identified: electron transfer from the anion to the probe or heavy

atom quenching. The enhanced sensitivity of the sensor to the iodide and bromide

ions over chloride supports a heavy atom perturbation quenching mechanism, but

the higher oxidation potentials of I– and Br– mean that they are also more likely to

induce alternative competing processes such as photoinduced electron transfer

(probably the quenching mechanism for pseudo-halides such as SCN–). In a study

using the probe N-methylquinolinium (Fig 12.3), analysis of the oxidation

potentials for each quencher revealed a linear relationship between the quenching

efficiency and the oxidation potential, suggesting that electron transfer was the

predominant quenching mechanism [20]. Efficient quenching by I– and Br– and

weaker quenching by Cl– makes it particularly challenging to develop selective

optical sensors for chloride, which is an important biological analyte due to its role

in cellular pH, fluid adsorption and neuronal transmission processes. Photoinduced Electron Transfer

PET involves the transfer of an electron from a donor, D, to an acceptor, A,

following the absorption of light. The direction of electron transfer is determined

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

2…Optical Properties and Their Exploitation in Sensing

Tải bản đầy đủ ngay(0 tr)