Transcript Chemical Sensors - University of Akron

Chemical Sensors
Chapter 8
Introduction







Chemical sensors are very different
Sensing is usually based on sampling
Sample is allowed to react in some fashion with
elements of the sensor
Usually an electric output is produced
Transduction can be multi-stage and complex
In some sensors, a complete analysis of the
substance occurs
In others a direct output occurs simply due to the
presence of the substance.
Introduction





Chemical sensing is quite common
Used in industry for process control and for
monitoring, including monitoring for safety.
Important role in environmental protection
Tracking of hazardous materials
Tracking natural and man made occurrences




pollution,
waterways infestation
migration of species
weather prediction and tracking.
Introduction




In sciences and in medicine - sampling of
substances such as oxygen, blood, alcohol
In the food industry for monitoring food safety
Military has been using chemical sensors at
least since WWI to track chemical agents used
in chemical warfare
Around the home and for hobbies (CO detection,
smoke alarms, pH meters)
Classification



Direct and indirect output sensors
Direct sensor: the chemical reaction or the
presence of a chemical produces a measured
electrical output.
Example: the capacitive moisture sensor – the
capacitance of a capacitor is directly
proportional to the amount of water present
between its plates.
Classification




Indirect (also called complex) sensor relies on a
secondary, indirect reading of the sensed stimulus.
Example: optical smoke detector. An optical sensor such
as a photoresistor is illuminated by a source and
establishes a background reading.
Smoke is “sampled” by allowing it to flow between the
source and sensor and alter the light intensity, its
velocity, its phase or some other measurable property.
Some chemical sensors are much more complex than
this and may involve more transduction steps. In fact,
some may be viewed as complete instruments or
processes.
Approach





Avoid a rigid classification
Concentrate on chemical sensors that are most
important from a practical point of view while
Try to cover most principles involved
Steer clear of most chemical reactions and the
formulas associated with them,
Replace these by physical explanations that
convey the process and explain the results
without the need for analytic chemistry.
Approach

Will start with the class of electrochemical sensors.


The second group studied are those sensors that
generate heat and the heat is the sensed quantity.



Includes those sensors that convert a chemical quantity directly
into an electrical reading and follows the definition above for
direct sensors.
These sensors, just like the thermo-optical sensors in chapter 4
are indirect sensors as are the optical chemical sensors.
Following these are some of the most common sensors
such as pH and gas sensors.
Humidity and moisture sensors are included here even
though their sensing is not truly chemical but because
the sensing methods and materials relate to chemical
sensors.
Electrochemical sensors





Expected to exhibit changes in resistance
(conductivity) or changes in capacitance
(permittivity) due to substances or reactions.
These may carry different names.
Potentiometric sensors do not involve current –
measurement of capacitance and voltage.
Amperimetric sensors rely on measuring current
Conductimetric sensors rely on measurement of
conductivity (resistance).
Electrochemical sensors


These are different names for the same
properties since voltage, current and resistance
are related by Ohm’s law.
Electrochemical sensors include a large number
of sensing methods, all based on the broad area
of electrochemistry. Many common sensors
including fuel cells, surface conductivity sensors,
enzyme electrodes, oxidation sensors and
humidity sensors belong to this category.
Metal-oxide sensors



Rely on a very well known property of metal
oxides at elevated temperature to change their
surface potential, and therefore their conductivity
in the presence of various reducible gases such
as ethyl alcohol, methane and many other
gases, sometimes selectively sometimes not.
Metal oxides that can used are oxides of tin
(SnO2), zinc (ZnO), iron (Fe2O2), zirconium
(ZrO2), titanium (TiO2) and Wolfram (WO3).
These are semiconductor materials and may be
either p or n type (with preference to n –type).
Metal-oxide sensors




Fabrication is relatively simple
May be based on silicon processes or other thin
or thick film technologies.
The basic principle is that when an oxide is held
at elevated temperatures, the surrounding gases
react with the oxygen in the oxide causing
changes in the resistivity of the material.
The essential components are the high
temperature, the oxide and the reaction in the
oxide
Metal-oxide sensors

Typical sensor: CO sensor shown in Figure
8.1a.

Consists of a heater and a thin layer of SnO2
Metal-oxide sensors







Construction:
A silicon layer is first created to serve as temporary
support for the structure.
Above it an SiO2 layer is thermally grown.
This layer can withstand high temperatures.
On this a layer of gold is sputtered and etched to form a
long meandering wire.
The wire serves as the heating element by driving it with
a sufficiently high current.
A second layer of SiO2 is deposited.
Metal-oxide sensors




Then the SnO2 oxide is sputtered on top and patterned
with grooves on top to increase its active surface.
The original silicon material is etched away to decrease
the heat capacity of the sensor.
The sensing area can be quite small – 1-1.5 mm2.
The device is heated to 300 C to operate but, because
the size is very small and the heat capacity small as well,
the power needed is typically small, perhaps of the order
of 100 mW.
Metal-oxide sensors - operation

Conductivity of the oxide can be written as:
 = 0 + kPm
0 is the conductivity of the tin oxide at 300C, without
CO present
P is the concentration of the CO gas in ppm (parts per
million),
k is a sensitivity coefficient (determined experimentally
for various oxides)
m is an experimental value - about 0.5 for tin oxide.
Metal-oxide sensors - operation


Conductivity increases with increase in
concentration as shown in Figure 8.1b.
Resistance is proportional to the inverse of
conductivity so that it may be written as
R = aP a
a is a constant defined by the material and construction
and
a an experimental quantity for the gas.
P is the concentration.
Response of a metal-oxide
sensor
Metal-oxide sensors - operation



The response is exponential (linear on the log
scale)
A transfer function of the type shown in
Figure 8.1b must be defined for each gas
and each type of oxide.
SiO2 based sensors as well as ZnO sensors
can also be used to sense CO2, touluene,
benzene, ether, ethyl alcohol and propane
with excellent sensitivity (1-50ppm).
Metal-oxide sensors - Variations





A variation of the structure above is shown in
Figure 8.2.
It consists of an SnO2 layer on a ferrite
substrate.
The heater here is provided by a thick layer of
RuO2, fed through two gold contacts.
The resistance of the very thin SnO2 (less
than about 0.5 m) is measured between two
gold contacts.
This sensor, which operates as described
previously is sensitive to ethanol and carbon
monoxide
Ethanol/ CO sensor
Metal-oxide sensors - notes



The reaction is with oxygen
Any reducible gas (a gas that reacts with
oxygen) will be detected.
Lack of selectivity - common problem in metal
oxide sensors. To overcome it,



Select temperatures at which the required gas reacts
The particular gas may be filtered.
These sensors are used in many applications
form CO and CO2 detectors to oxygen sensors
in automobiles.
Metal-oxide sensors - notes




Example: oxygen sensors in automobiles use a TiO2
sensor built as above in which resistance increases in
proportion to the concentration of oxygen.
This is commonly used in other application such as
oxygen in water (for pollution control purposes).
The process can also be used to determine the amount
of available organic material in water by first evaporating
the water and then oxygenating the residue to determine
how much oxygen is consumed using an oxygen sensor.
The amount of oxygen is then an indication of the
amount of organic material in the sample.
Solid elecrolyte sensor




Another important type of sensor is the solid
electrolyte sensor
Has found significant commercial application
Most often used in oxygen sensors, including
those in automobiles.
Principle: a galvanic cell (battery cell) is built
which produces an emf across two electrodes
based on the oxygen concentrations at the two
electrodes under constant temperature and
pressures.
Solid elecrolyte sensor





A solid electrolyte capable of operating at high
temperatures is used
Usually made of zirconium dioxide (ZrO2) and Calcium
oxide (CaO) in a roughly 90% 10% ratio
It has high oxygen ion conductivity at elevated
temperatures (above 500C).
The electrolite is made of sintered ZrO2/ CaO powder
which makes it into a ceramic material.
The inner and outer electrodes are made of platinum
which act as catalysts and absorb oxygen. The structure
is shown in Figure 8.3 for an exhaust oxygen sensor in a
car engine.
Solid electrolyte oxygen sensor
Solid electrolyte sensor operation

The potential across the electrodes is
1
pO2
RT
em f =
ln
2
4F pO2
R is the gas constant (=8.314 J/K/mol),
T is the temperature (K)
F is the Faraday constant (=96487 C/mol).
P1 is the concentration of oxygen in the exhaust,
P2 the concentration of oxygen in the atmosphere, both
heated to the same temperature.
Solid electrolyte sensor - use



Used to adjust the fuel ratio at the most efficient
rate at which pollutants (NOx and CO) are
converted into nitrogen (N2), carbon dioxide
(CO2) and water (H2O), all of which are natural
constituents in the atmosphere and hence
considered non-pollutants
Usually fuel is enriched to achieve full
combustion of pollutants
A PASSIVE SENSOR!
Solid electrolyte sensor - use as
an active sensor



Many engines operate in a much leaner mode
(for better fuel efficiency),
The solid electrolyte sensor is not sufficiently
sensitive (the amount of oxygen in the exhaust is
high and the reading of the electrolytic cell is
insufficient).
The solid electrolyte sensor is modified to act as
a passive sensor
Solid electrolyte sensor - use as
a passive sensor






A solid electrolyte between two platinum electrodes as
shown in Figure 8.4. are used, but:
A potential is applied to the cell.
This arrangement forces (pumps) oxygen across the
electrolyte and a current is produced proportional to the
oxygen concentration in the exhaust.
The current is then a measure of the oxygen
concentration in the exhaust
This sensor is called a diffusion oxygen sensor or the
diffusion-controlled limiting current oxygen sensor.
Operates similar to charging a battery
Diffusion-controlled current
limiting oxygen sensor
Oxygen sensor for molten metal






Important in oxygen sensing in production of steel and
other molten materials
The quality of the final product is a direct result of the
oxygen in the process. The sensor is shown in Figure
8.5.
The molybdenum needle keeps the device from melting
when inserted in the molten steel.
A potential difference is developed across the cell
(between the molybdenum and the outer layer).
The voltage is measured between the inner electrode
and outer layer through an iron electrode dipped into the
molten steel.
The voltage developed is directly proportional to the
oxygen concentration in the molten steel.
Oxygen sensor for molten
metals
The MOS chemical sensor




Use of the basic MOSFET structure commonly
used in electronics, as a chemical sensor.
The basic idea: the classical MOSFET transistor
in which the gate serves as the sensing surface.
Advantage: a very simple and sensitive device is
obtained which controls the current through the
MOSFET.
The interfacing of such a device is simple and
there are fewer problems (such as heating,
temperature sensing, etc.) to overcome.
MOS chemical sensors

Example, by simply replacing the metal
gate in Figure 8.6 with palladium, the
MOSFET becomes a hydrogen sensor
MOS chemical sensors





Palladium absorbs hydrogen and its potential changes
accordingly.
Sensitivity is down to about 1 ppm.
Similar structures can sense gases such as H2S and
NH3.
Palladium mosfets (Pd-gate MOSFET) can also be used
to measure oxygen in water, relying on the fact that the
absorption efficiency of oxygen goes down in proportion
to the amount of oxygen present.
We shall say much more about the MOSFET sensor in
the subsequent section on PH sensing since these have
been very successful in this capacity.
Potentiometric sensors




A large subset of electrochemical sensors
Principle: electric potential develops at the
surface of a solid material immersed in solution
containing ions that exchange at the surface.
The potential is proportional to the number or
density of ions in the solution.
A potential difference between the surface of the
solid and the solution occurs because of charge
separation at the surface.
Potentiometric sensors



The contact potential, analogous to that used to set up a
voltaic cell cannot be measured directly.
If a second electrode is provided, an electrochemical cell
is setup and the potential across the two electrodes is
directly measurable.
To ensure that the potential is measured accurately, and
therefore that the ion concentration is properly
represented by the potential, it is critical that the current
drawn by the measuring instrument is as small as
possible (any current is a load on the cell and therefore
reduces the measured potential).
Potentiometric sensors




For a sensor of this type to be useful, the
potential generated must be ion specific – that
is, the electrodes must be able to distinguish
between solutions.
These are called ion-specific electrodes or
membranes.
The four types of membranes are:
Glass membranes, selective for H+, Na+ and
NH4+ and similar ions.
Potentiometric sensors



Polymer-immobilized membranes: In this type of
membrane, an ion-selective agent is immobilized
(trapped) in a polymer matrix. A typical polymer is PVC
Gel-immobilized enzyme membranes: the surface
reaction is between an ion specific enzyme which in turn
is either bonded onto a solid surface or immobilized into
a matrix - mostly for biomedical applications
Soluble inorganic salt membranes: either crystalline or
powdered salts pressed into a solid are used. Typical
salts are LaF3 or mixtures of salt such as Ag2S and AgCl.
These electrodes are selective to F, S and Cl and
similar ions.
Glass membrane sensors






By far the oldest of the ion-selective electrodes,
Used for pH sensing from the mid-1930’s and is
as common as ever.
The electrode is a glass made with the addition
of sodium (Na2O) and aluminum oxide (Al2O3),
Made into a very thin tube-like membrane.
This results in a high resistance membrane
which nevertheless allows transfer of ions
across it.
The basic method of pH sensing is shown in
Figure 8.7a.
pH sensor
pH sensor





Consists of the glass membrane electrode on the left
and a reference electrode on the right.
The reference electrode is typically an Ag/AgCl
electrode in a KCl aqueous solution or a saturated
Calomel electrode (Hg/Hg2Cl2 in a KCl solution).
The reference electrode is normally incorporated into
the test electrode so that the user only has to deal
with a single probe as shown in Figure 8.7b.
The sensor is used by first immersing the electrode
into a conditioning solution of Hcl (0.1.mol/liter) and
then immersing it into the solution to be tested. The
electric output is calibrated in pH.
A sensor of this type responds to pH from 1 to 14.
pH probe with reference
electrode
Glass membrane sensors

Modifications of the basic configuration, both in
terms of the reference electrode (filling) as well
as the constituents of the glass membrane lead
to sensitivity to other types of ions as well as to
sensors capable of sensing dissolved gas in
solutions, particularly ammonia but also CO2,
SO2, HF, H2S and HCN
Soluble inorganic salt
membrane sensors



Based on soluble inorganic salts which undergo
ion-exchange interaction in water and generate
the required potential at the interface.
Typical salts are the lanthanum fluoride (LaF3)
and silver sulfide (Ag2S).
The membrane may be either




a singe crystal membrane,
a sintered disk made of powdered salt
a polymer matrix embedding the powdered salt
each has its own application and properties
Soluble inorganic salt
membrane sensors





The structure of a commercial sensor used to
sense fluoride concentration in water is shown
next
The sensing membrane, made in the form of a
thin disk grown as a single crystal.
The reference electrode is created in the internal
solution (in the case: NaF/NaCl at 0.1 mol/liter).
The sensor shown can detect concentrations of
fluoride in water between 0.1 and 2000 mg/l.
This sensor is commonly used to monitor
fluoride in drinking water (about 1mg/l).
Soluble inorganic salt
membrane sensors for fluoride
Soluble inorganic salt
membrane sensors




Membranes may be made of other materials
such as silver sulfide.
The latter is easily made into thin sintered disks
from powdered material and may be used in lieu
of the single crystal.
Other compounds may be added to affect the
properties of the membrane and hence
sensitivities to other ions.
This leads to selective sensors sensitive to ions
of chlorine, cadmium, lead and copper and are
often used to sense for dissolved heavy metals
in water.
Polymeric salt membranes




Polymeric membranes are made by use of
a polymeric binder for the powdered salt
About 50% salt and 50% binding material.
The common binding materials are PVC,
polyethylene and silicon rubber.
In terms of performance these membranes
are quite similar to sintered disks.
Polymer-immobilized
ionophore membranes




A development of the inorganic salt membrane
Ion-selective, organic reagents are used in the
production of the polymer by including them in
the plasticizers, particularly for PVC.
A reagent, called ionophore (or ion-exchanger)
is dissolved in the plasticizer (about 1% of the
plasticizer).
This produces a polymer film which can then be
used as the membrane replacing the crystal or
disk in sensors.
Polymer-immobilized ionophore
membranes



The construction of the
sensor is simple
Shown in Figure 8.9 and
includes an Ag/AgCl
reference electrode.
The resulting sensor is a
fairly high resistance
sensor.
Polymer-immobilized ionophore
membranes





A different approach to building
polymer-immobilized ionophore
membranes is shown in Figure
8.10.
It is made of an inner platinum wire
on which the polymer membrane is
coated
The wire is protected with a coating
of paraffin.
This is called a coated wire
electrode.
To be useful a reference membrane
must be added.
Gel-immobilized enzyme
membranes




Similar in principle to polymer immobilized
ionophore membranes
A gel is used and the ionophore is replaced by
an enzyme which is selective to a particular ion.
The enzyme, (a biomaterial) is immobilized in a
gel (polyacrylamide) and held in place on a
glass membrane electrode as shown in Figure
8.11.
The choice of the enzyme and the choice of the
glass electrode define the selectivity of the
sensor.
Gel-immobilized enzyme
membrane sensor
Gel-immobilized enzyme
membrane sensors




Gel sensors exist for the sensing of a variety of
important analytes including urea glucose, Lamino acids, penicillin and others.
The operation is simple; the sensor is placed in
the solution to be sensed which diffuses into the
gel and reacts with the enzyme.
The ions released are then sensed by the glass
electrode.
These sensors are slow in response because of
the need for diffusion but they are very useful in
analysis in medicine including blood and urine.
The Ion-sensitive field effect
transistor ISFET




Also called the ChemFet
Essentially a MOSFET in which the gate has
been replaced by an ion-selective membrane.
Any of the membranes above may be used most often the glass and polymeric membranes
In its simplest form, a separate reference
electrode is used but the reference electrode
may be easily incorporated within the gate
structure as shown in Figure 8.12.
Ion-sensitive field effect
transistor ISFET
Ion-sensitive field effect
transistor ISFET




The gate is then allowed to come in
contact with the sample to be tested
The drain current is measured to indicate
the ion concentration.
The most important use of this device is
measurement of pH
Available commercially.
Thermo-chemical sensors


A class of sensors that rely on the heat
generated in chemical reactions to sense the
amount of particular substances (reactants).
There are three sensing strategies, each leading
to sensors for different applications.



sense the temperature rise due to the reaction
catalytic sensor used for sensing of flammable gases.
measures the thermal conductivity in air due to the
presence of a sensed gas.
Thermisotor based chemical
sensors




Principle: sense the small change in
temperature due to the chemical reaction.
A reference temperature sensor is usually
employed to sense the temperature of the
solution
The difference in temperature is then related to
the concentration of the senses substance.
The most common approach is to use an
enzyme based reaction (enzymes are highly
selective - so that the reaction can be
ascertained - and because they generate
significant amounts of heat).
Thermisotor based chemical
sensors



A typical sensor is made by
coating the enzyme directly
on the thermistor.
The thermistor itself is a
bead thermistor which
makes for a very compact
highly sensitive sensor.
The construction is shown in
Figure 8.13.
Thermisotor based chemical
sensors





Used to sense concentration of urea and glucose, each
with its own enzyme (urease or glucose enzymes).
The amount of heat generated is proportional to the
amount of the substance sensed in the solution.
The temperature difference between the treated
thermistor and the reference thermistor is then related to
the concentration of the substance.
A thermistor can measure temperature differences as
low as 0.001C but most are less sensitive than that
Overall sensitivity depends on the amount of heat
generated.
Catalytic sensors





True calorimetric sensors:
A sample of the (gas) analyte is burned
The heat generated in the processed is
measured through a temperature sensor.
This type of sensor is very common
Main tool in detection of flammable gases such
as methane, butane, carbon monoxide and
hydrogen, fuel vapors such as gasoline as well
as flammable solvents (ether, acetone, etc.).
Catalytic sensors






Principle: sampling of air containing the flammable gas
into a heated chamber
Combusts the gas to generate heat.
To speed up the process, a catalyst is used.
The temperature sensed is then indicated as a
percentage of flammable gas in air.
The simplest form of a sensor is to use a platinum coil
through which a current is passed.
The platinum coil heats up due to its own resistance and
serves as a catalyst for hydrocarbons (this is the reason
why it is the active material in a catalytic converter in
automobiles).
Catalytic sensors


The released heat raises the temperature of the
coil.
This resistance is then a direct indication of the
amount of flammable gas in the sampled air.
Catalytic sensors




Better catalysts are palladium and rhodium
One such sensor, called a “pellistor” (the name
comes from Pellister who discovered the
process), is shown in Figure 8.14.
It uses the same heater and temperature
sensing mechanism (platinum coil)
Uses a palladium catalyst either external to the
ceramic bead or embedded in it.
Catalytic sensor (pelistor)
using a catalyst layer
Catalytic sensors




The second is better because there is less of a
chance of contamination by noncombustible
gases (called poisoning – which reduce
sensitivity).
Advantage: operate at lower temperatures
(about 500C as opposed to about 1000C for
the platinum coil sensor).
A sensor of this type will contain two beads, one
inert (serving as reference) and one sensing
bead, in a common sensing head shown in
Figure 8.15.
This generates a reaction in a few seconds.
Catalytic sensors with
reference pelistor
Catalytic sensors application





Used in mines to detect methane and in industry
to sense solvents in air.
The most important issue is the concentration at
which a flammable gas explodes.
This is called the lower explosive limit (LEL),
below which a gas will not ignite.
For methane for example, the LEL limit is 5% (by
volume, in air).
A methane sensor will be calibrated as % of LEL
(100%LEL corresponds to 5% methane in air)
Thermal conductivity sensor






Does not involve any chemical reaction
Uses the thermal properties of gases for detection.
A sensor of this type is shown in Figure 8.16.
It consists of a heater set at a given temperature (around
250C).
The heater looses heat to the surrounding area,
depending on the gas with which it comes in contact.
As the gas concentration becomes higher a larger
amount of heat is lost compared to loss in air and the
temperature of the heater as well as its resistance
diminish.
Thermal conductivity sensor
Thermal conductivity sensor




This change in resistance is sensed and
calibrated in terms of gas concentration.
Unlike the previous two types of sensors, this
sensor is useful for high concentrations of gas.
It can be used for inert gasses such as nitrogen,
argon and carbon dioxide as well as for volatile
gases.
The sensor is in common use in industry and is
a useful tool in gas chromatography in the lab.
Optical chemical sensors



Transmission, reflection and absorption
(attenuation) of light in a medium, its velocity
and hence its wavelength are all dependent of
the properties of the medium.
These can all serve as the basis of sensing
either by themselves or in conjunction with other
transduction mechanisms and sensors.
For example, the optical smoke detector uses
the transmission of light through smoke to detect
the presence of smoke.
Optical chemical sensors





Other substances are sensed in this way, sometimes by
adding agents to, for example, color the substance
tested.
More complex mechanisms are used to obtain highly
sensitive sensors to a variety of chemical conditions.
In many optical sensors, use is being made of an
electrode which, when in the substance being tested,
changes some optical property of the electrode.
An electrode of this type is called an “optode” in parallel
with “electrode”.
The optode has an important advantage in that no
reference is needed and it is well suited for use with
optical guiding systems such as optical fiber.
Optical chemical sensors




Other options for opto-chemical sensing are the
properties of some substances to fluoresce or
phosphoresce under optical radiation.
These chemiluminescence properties can be senses and
used for indication of specific materials or properties.
Luminescence can be a highly sensitive method
because the luminescence is at a different frequency
(wavelength) than the frequency (wavelength) of the
exciting radiation.
This occurs more often with UV radiation but can occur
in the IR or visible range as well and is often used for
detection.
Optical chemical sensors


Optical sensing mechanisms rely at least in part on
absorption of light by the substance through which it
propagates or on which it impinges.
This absorption, is governed by the Beer-Lambert
law, stated as follows:
A = bM
 is the absorption coefficient characteristic of the medium
[103cm2/mol],
b is the path length [cm] traveled and
M is the concentration in [mol/l].
A=log(P0/P) is the absorbance where P0 is the incident and P the
transmitted light intensity.
Optical chemical sensor





The simplest sensors are the reflectance
sensors
Rely on the reflective properties of a membrane
or substance to infer a property of the
substance.
In many of these sensor a fiber optic cable or an
optical waveguide are used.
The basic structure is shown in Figure 8.23.
A source of light (LED, white light, laser)
generates a beam which is conducted through
the optical fiber to the optode.
Reflection optical sensor
Optical chemical sensor





The optical properties of the optode are altered by the
substance to which it reacts
The reflected beam is then a function of the
concentration of the analyte or its reaction products in
the optode.
It is also possible to separate the incident and reflected
beams by separate optical guides but usually this is not
necessary.
An alternative way of sensing is to use an uncladded
optical fiber so the light is lost through the walls of the
fiber.
This is called an evanescent loss and depends on what
is in contact with the walls of the fiber.
Evenescent field sensing
Optical chemical sensor



In this type of sensor the coupling to the optode
is through he walls of the fiber rather than its
end.
This also means that rather than reflection, the
transmission through the fiber is measured.
The transmitted wave is then dependent on the
amount of light absorbed in the optode and
therefore a function of the analyte in the optode.
Optical pH sensor




pH sensing can be done optically by using
special optodes which change color with change
in pH.
In these systems, only about one pH unit on
either side of the pH of the optode (before the
analyte interaction) can be sensed.
This span is sufficient for some applications in
which the range is narrow.
A sensor of this type is shown in Figure 8.25.
Reflection pH sensor
Optical pH sensor







A hydrogen permeable membrane is used in which
phenol red is immobilized on polyacrylamide
microspheres.
The membrane is a dialysis tube (cellulose acetate)
The optode is attached to the end of an optical fiber.
When immersed the analyte, diffuses into the optode.
Phenol red is known to absorb light at a wavelength of
560 nm (yellow-green light).
The amount of light absorbed depends on pH and hence
the reflected light will change with pH.
The difference between the incident and reflected
intensities is then related to pH.
Optical pH sensor




A similar sensor uses the fluorescent properties
of HPTS (a weak acid).
This substance fluoresces when excited by UV
light at 405 nm.
The intensity of fluorescence is then related to
the pH.
This material is particularly useful since its
normal pH is 7.3 so that measurements around
the neutral point can be made and in particular
in physiological measurements.
Optical pH sensor





Optodes can also be used to sense ions.
Metal ions are particularly easy to sense because they
can form highly colored complexes with a variety of
reagents.
These reagents are embedded in the optode and the
reflectivity properties are then related to concentration of
the metal ions.
Fluorescence is also common in metal ions, a method
that is used extensively in analytical chemistry, primarily
by use of UV light, with fluorescence in the visible range.
These methods have been used to sense a variety of
other ions including oxygen in water, penicillin and
glucose in blood and others.
Mass sensors





Detect the changes in the mass of a sensing
element due to absorption of an analyte.
Masses involved in absorption are minute
A method must be found that will be sensitive to
these minute mass changes.
Mass sensors are also called microgravimetric
sensors.
In a practical sensor it is not possible to sense
this change in mass and therefore indirect
methods must be used.
Mass sensors





This is done by using piezoelectric crystals such as
quartz
Setting them into oscillation at their resonant frequency
(see chapter 7).
This resonant frequency is dependent on the way the
crystal is cut and on dimensions but once these have
been fixed, any change in mass of the crystal will change
its resonant frequency.
The sensitivity is generally very high - of the order of 10
g/Hz and a limit sensitivity of about 10g.
Since the resonant frequency of crystals can be very
high, the change in frequency due to change in mass is
significant and can be accurately measured digitally.
Mass sensors



An equivalent approach can be taken with SAW
resonators which,
They can resonate at even higher frequencies than
crystals and hence offer higher sensitivities.
The shift in resonant frequency can be written as:
f = f0 Smm
f0 is the base resonant frequency
Sm is a sensitivity factor that depends of the crystal (cut,
shape, mounting, etc.)
m is the change in mass.
Mass sensors




The mass due to the analyte may be
absorbed directly into the crystal (or any
piezoelectric material) or in a coating on the
crystal.
Simple and efficient sensors.
Selectivity is poor since crystals and coatings
tend to absorb more than one species
confounding discrimination between species.
A basic requirement is that the process be
reversible, that is, the absorbed species must
be removable (by heating) without any
hysteresis.
Mass sensors - humidity
sensing






The most common analyte is water vapor
A mass humidity sensor is made by coating the
crystal by a thin layer of hygroscopic material
There are many hygroscopic materials that may be
used including polymers, gelatins, silica, fluorides.
The moisture is removed after sensing by heating.
A sensor of this type can be quite sensitive but its
response time is slow.
It may take many seconds (20-30sec) for sensing
and many more for regeneration (30-50 sec).
Mass sensors - notes



The method is very useful and has been
applied to sensing of a large variety of gases
and vapors, some being sensed at room
temperatures, some at elevated
temperatures.
The main difference between sensing one
gas or another is in the coating, in an attempt
to make the sensor selective.
The applications are mostly in sensing of
noxious gases and in dangerous substances
such as mercury.
Mass sensors - notes





Sensing of sulfur dioxide (mostly due to burning of coal and
fuels) is by amine coatings which react with sulfur dioxide.
Sensitivities as low as 10 ppb are detectable.
When detecting ammonia (for application in environmental
effects of waste water and sewage), the coating is ascorbic acid
or pyridoxine hydrochloride (and some similar compounds) with
sensitivities down to micrograms/kg.
Hydrocarbon sulfide is similarly detected by using acetate
coatings (silver, copper, lead acetates as well as as others).
Mercury vapor is sensed by the use of gold as a coating since
the two elements form an amalgalm that increases the mass of
the gold coating.
Other applications are in sensing hydrocarbons, nitro-toluenes
(emitted by explosives) and gases emitted by pesticides,
insecticides and other sources.
SAW mass sensors




A SAW mass sensor is made as a delay line
resonator, as we have seen in chapter 7.
The delay line itself is now coated with the
specific reactive coating for the gas to be
sensed.
This is shown in Figure 8.17.
To operate, air containing the gas is sampled
(drawn above the membrane) and the resonant
frequency measured.
SAW mass sensor
SAW mass sensors




Can be used to sense solid particles such as pollen or
pollutants by replacing the membrane with a sticky
substance.
The problem then would be the regeneration – cleaning
the surface for the next sampling.
The choice of coating determines the selectivity of the
sensor. Table 8.1 shows some sensed substances and
the appropriate coatings.
Sensitivities of saw resonators can be much higher than
crystal resonators with limit sensitivities of approximately
10-15g. Sensitivities expected are of the order of 50
Hz/Hz. (25 kHz shift for a 500 Mhz resonator)
Coatings and analytes for SAW
sensors
Ta ble 8.1. Some sensed su bstances an d the coati ngs used for th at purpose .
Compound
Chemical coat ing
SAW mterial
SO2
TEA
Lithium Niobate
H2
Pd
Lithium Niobate, Silicon
NH3
Pt
Quartz
H2S
WO3
Lithium Niobate
Water vapor
Hygroscopic material
Lithium Niobate
NO2
PC
Lithium Niobate, Qu7artz
NO2, NH3, SO2, CH4
PC
Lithium Niobate
Explosives vapor, drug s
Polymer
Quartz
SO2, methane
none
Lithium Niobate
TEA=T rithanolamine, PC=Phthalocyamine.
Humidity and moisture sensors




SAW sensors is indicated are common sensors
There are however other methods of sensing
humidity
All involve some type of hygroscopic medium to
absorb water vapor.
These can take many forms - capacitive,
conductive and optical are the most common
Humidity and moisture sensors




The terms humidity and moisture are not
interchangeable.
Humidity refers to the water content in gases such as in
the atmosphere.
Moisture is the water content in any solid or liquid.
Other important, related quantities are




dew point temperature
absolute humidity and
relative humidity.
These are defined as follows:
Humidity and moisture sensors



Relative humidity is the ratio of the water
vapor pressure of the gas (usually air) to the
maximum saturation water vapor pressure in
the same gas at the same temperature.
Saturation is that water vapor pressure at
which droplets form. The atmospheric
pressure is the sum of the water vapor
pressure and the dry air pressure.
Relative humidity is not used above the
boiling point of water (100C) since the
maximum saturation above that temperature
changes with temperature.
Humidity and moisture sensors


Dew-point temperature is the temperature at
which relative humidity is 100%. This is the
temperature at which air can hold maximum
amount of moisture. Cooling below it creates
fog (water droplets), dew or frost.
Absolute humidity is defined as the mass of
water vapor per unit volume of wet gas in
grams/cubic meter [g/m3].
Humidity and moisture sensors




The simplest moisture sensor is capacitive
sensor
It relies on the change in permittivity due to
moisture.
The permittivity of water is rather high (800 at
low frequencies).
Humidity of course is different than liquid
water and hence the permittivity of humid air
is either given in tables as a function of
relative humidity or may be calculated from
the following empirical relation:
Humidity and moisture sensors
 = 1 + 211 P + 48Ps H 106 0
T





T
0 is permittivity of vacuum,
T is the absolute temperature [K],
P is the pressure of moist air [mm Hg],
H is the relative humidity [%]
Ps is the pressure of saturated water vapor
at the temperature T [mm Hg]
Humidity and moisture sensors


The capacitance of a parallel plate capacitor is
C=A/d
This establishes a relation between capacitance
and relative humidity:
C = C0 + C0211P106 + C0
T
21148Ps
T2
106H
C0 is the capacitance of the capacitor in vacuum
(C0=A/d).
This relation is linear at any given pressure and temperature.
Humidity and moisture sensors



In more practical designs, means of
increasing this capacitance are used.
Use a hygroscopic material between the
plates both to increase the capacitance at no
humidity and to absorb the water vapor.
(hygroscopic polymer films.
The metal plates are made of gold. In a
device of this type the capacitance is
approximated as:
C = C0 + C0 ah H
where ah is a moisture coefficient
Humidity and moisture sensors




Method assumes that the moisture content in the
hygroscopic polymer is directly proportional to
relative humidity and that
As the humidity changes, the moisture content
changes (that is, the film does not retain water).
Under these conditions the sensing is
continuous but, as expected, changes are slow
and
A sensor of this type can sense relative humidity
from about 5% to 90% at an accuracy of 2-3%.
Humidity and moisture sensors





In a parallel plate capacitor the film must be thin
Moisture can only penetrate from the sides.
It is therefore slow to respond to changes in
moisture because of the time it takes for
moisture to penetrate throughout the film.
A different approach is shown in Figure 8.18.
Here the capacitor is flat and built from a series
of interdigitated electrodes to increase
capacitance.
Capacitive moisture sensor
Capacitive moisture sensors






The hygroscopic dielectric may be made of SiO2 or
phosphorosilicate glass.
The layer is very thin to improve response.
Because the sensor is based on silicon, temperature
sensors are easily incorporated as are other
components such as oscillators.
The capacitance of the device is low and therefore it
will be used as part of an oscillator and the frequency
sensed.
However, the permittivity of the dielectric is frequency
dependent (goes down with frequency).
This means that frequency cannot be too high,
especially if low humidity levels are sensed.
Resistive moisture sensors





Humidity is known to change the resistivity
(conductivity) of some nonconducting
materials.
This can be used to build a resistive sensor.
A hygroscopic conducting layer and two
electrodes are provided.
The electrodes will be interdigitated to
increase the contact area, as shown in
Figure 8.19.
The hygroscopic conductive layer must have
a relatively high resistance which goes down
with humidity (actually absorbed moisture).
Resistive moisture sensor
Resistive moisture sensors





Materials that can be used for this purpose include
polystyrene treated with sulfuric acid and solid
polyelectrolytes
A better structure is shown in Figure 8.20. It operates
as above but the base material is silicon.
An aluminum layer is formed on the silicon (highly
doped so its resistivity is low).
The aluminum layer is oxidized to form a layer of
Aluminum oxide which is porous and hygroscopic.
An electrode of porous gold is deposited on top to
create the second contact and to allow moisture
absorption in the Al2O3 layer
Thermally conductive moisture
sensor
Thermally conductive
moisture sensors




Humidity may also be measured through thermal
conduction
Higher humidity increases thermal conduction.
This sensor however senses absolute humidity
rather than relative humidity.
The sensor makes use of two thermistors
connected in a differential or bridge connection
(bridge connection is shown in Figure 8.21.
Thermally conductive moisture
sensor
Thermally conductive moisture
sensor





The thermistors are heated to an identical temperature
by the current through them so that the output is zero in
dry air.
One thermistor is kept in an enclosed chamber as a
reference and its resistance is constant.
The other is exposed to air and its temperature changes
with humidity.
As humidity increases, its temperature drops and hence
its resistance increases (for NTC thermistors).
At saturation the peak is reached. Above that the output
drops again (Figure 8.21b).
Optical humidity sensor





By measuring the ambient temperature t, and
then evaluating the dew point temperature DPT,
RH is calculated from Eq. (8.1).
The basic idea is to use a dew point sensor.
The latter is built as shown in Figure 8.22.
The sensors is based on detecting the dew point
on the surface of a mirror.
To do so, light is reflected off the mirror and the
light intensity monitored.
Optical humidity sensor





A Peltier cell is used to cool the mirror to its dew point.
When the dew point temperature is reached, the
controller keeps the mirror at the dew point temperature.
The reflectivity now drops since water droplets form on
the mirror (the mirror fogs up).
This temperature is measured and is the dew point
temperature in Eq. (8.1).
Although this is a rather complex sensor and includes
the reference diodes for balancing, it is rather accurate,
capable of sensing the dew point temperature at
accuracies of less than 0.05C
Optical dew point temperature
sensor
Mass/SAW resonator dew point
temperature sensor





The same measurement can be done with the
mass sensor described in the previous section.
The resonant frequency of a crystal, covered with
a water-selective coating is used and its resonant
frequency sensed while the sensor is cooled.
At the dew point, the sensor’s coating is saturated
and the frequency is the lowest.
Equally well, a SAW mass sensor may be used
with even higher accuracy.
The heating/cooling is achieved as in Figure 9.22
by use of a Peltier cell.