Note: Descriptions are shown in the official language in which they were submitted.
CA 02391451 2008-03-31
REFERENCING AND RAPID SAMPLING IN ARTIFICIAL
OLFACTOMETRY
FIELD OF THE INVENTION
In general, this invention relates to chemical sensing, and in particular, to
referencing, rapid sampling and methods of reducing or eliminating drift in
artificial
olfactometry.
BACKGROUND OF THE INVENTION
An electronic nose is an array of chemical sensors coupled with
computerized multivariate statistical processing tools. These sensors respond
to a wide
variety of analytes giving rise to a unique signature or pattern for a given
analyte. The
pattern is interpreted using pattern recognition algorithms to identify or
quantify the
analyte of interest.
In general, the chemical sensors are based on physical or chemical
absorption, chemical desorption or optical properties that take place on the
sensors.
Suitable sensor types include metal oxide semiconductors, metal oxide
semiconducting
field effect transistors, conducting organic polymers, quartz microbalance,
surface
acoustic wave devices and conducting and nonconducting regions sensors.
For the analysis of organic solvent vapors, certain devices, such as
surface acoustic wave devices, respond to the extent of vapor sorption. This
sorption is
typically rapid, reversible and is proportional to vapor concentration.
However, various
drawbacks exist. For example, certain sensors are susceptible to humidity,
have low
confidence limits, are susceptible to drift and are unstable. In certain
instances,
instability can be corrected using background subtraction techniques. Humidity
in the
vapor can be eliminated by using a preconcentrator with water vapor
absorbents.
Confidence limits can be enhanced by using a limit of recognition that is
defined as the
1
CA 02391451 2009-01-16
concentration below which a vapor can no longer be reliably recognized from
its
response pattern (see, Zellers et al., Analytical Chemistry, 70, 4191-4201
(1998)).
Drift is one of the most serious drawbacks of sensor technology. Drift is
defined as the temporal shift of sensor response under constant or static
conditions.
The reason for certain types of drift is not well understood, but it is
believed to result
from unknown dynamic processes. Temperature or pressure fluctuations, or
changes
in the sensing environment can also cause drift. When the reasons for drift
are known,
it is sometimes possible to develop mathematical models that can compensate
for its
effects (see, Semin et al., Meas. Techn. 38, 30-32 (1995)). Work has been done
on
ways to improve the stability of sensors; however, it is not yet possible to
fabricate
sensors with no drift at all.
One possible solution to the effects of drift is to use a reference gas (see,
Fryder et al. Transducers `95 and Eurosensors IX., Stockholm Sweden, pp. 683-
686
(1995)). This technique is difficult or impractical in some situations, such
as a
handheld sensing device. Another technique is the use of a theory of hidden
variable
dynamics for the rejection of common mode drift. Moreover, the hidden variable
approach can be couple with adaptive estimation methods to compensate for
drift
(see, Holmberg et al., Sensors and Actuators B 42, 285-294 (1997).
In view of the inherent instability of certain sensor arrays, there remains a
need to have effective referencing and calibration in spite of the presence of
drift.
Devices and methods are needed which effectively produce reliable vapor
measurements in the presence of drift. The present invention fulfills these
and other
needs.
SUMMARY OF THE INVENTION
Temporal shift of sensor response under constant conditions is one of the most
serious drawbacks of sensor technology. Devices and methods are needed which
are
effective to produce reliable vapor measurements in the presence of temporal
shift.
As such, in certain aspects, the present invention provides a method for
reducing drift in an artificial olfaction device having an array of sensors.
The method
involves contacting the array of sensors with an analyte at a first
temperature
2
CA 02391451 2009-01-16
parameter to produce a first response, contacting the array of sensors with
the analyte
at a second temperature parameter to produce a second response, and measuring
a
differential response with respect to the first and second responses, then
referencing
the differential response to the first and second responses of the array of
sensors to
thereby reduce drift.
3
CA 02391451 2008-03-31
These and other aspects of the present invention will become more
apparent when read with the detailed description and figures that follow.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 illustrates a top perspective view of an embodiment of an
electronic nose device of the present invention.
Figure 2 illustrates a sensor response using a device of the present
invention.
Figure 3 illustrates a top perspective view of an embodiment of an
electronic nose device of the present invention.
Figure 4 illustrates differential flow pneumatics of the present invention.
Figure 5 illustrates a top sectional view of an embodiment of the present
invention.
Figure 6A-B illustrate Pane16A a top sectional view of an embodiment
of the present invention and Pane16B a module cover embodiment.
Figure 7 illustrates drift in sensor arrays.
Figure 8 illustrates a sensor response using a device of the present
invention.
DETAILED DESCRIPTION OF THE INVENTION AND PREFERRED
EMBODIMENTS
1. DEVICES
A. Externally Mounted Sensor Module
In certain aspects, the present invention provides a sensing device for
detecting an analyte, comprising: a housing; a sensor module mounted
externally on the
housing comprising an array of sensors, each of the sensors providing a
response in the
presence of the analyte; a monitoring device mounted on the housing and
configured to
monitor the responses of the array of sensors incorporated in the sensor
module, further
configured to produce a corresponding plurality of sensor signals; and an
analyzer
mounted on the housing and configured to analyze the plurality of sensor
signals to
identify the analyte.
4
CA 02391451 2008-03-31
Preferably, the sensor module is capable of automatic physical
movement. The physical movement is controlled by a controller that is
configured to
control the sensor module. This automatic physical movement of the sample
module
allows referencing of the sensor arrays. For instance, the sensor module is
placed in a
first position to be calibrated. The first position does not expose the sensor
array to the
test area or sample containing the analyte. Thus, the first position is a
calibration
position wherein the sensor response is set to a null value. After the sensor
array is
calibrated or nulled, the sensor module is automatically moved closer to the
test sample
or area containing the vapor or analyte to be measured. Sampling of the vapor
is quick
and reliable, and because the sensor module is externally mounted, the
response of the
sensor array is not impeded by sensor array pneumatics. As will be apparent to
those of
skill in the art, although the sensor array is set to a null value, it is
possible that some
background signal exists.
In an especially preferred embodiment, the externally mounted sensor is
on a handheld electronic nose. As shown in Figure 1, the sensor module is
externally
mounted. Figure 1 shows a top perspective view of an embodiment of a handheld
device. The handheld device includes an elongated housing having a lower end
sized to
be conveniently grasped and supported by the hand of an operator. A display 12
and
several push-button control switches 12a through 12c are located on the
housing's
topside, for convenient viewing and access by the operator. Push-buttons 12a-c
can be
used to control the device during its various operating modes. Display 12
displays
information about such operating modes and the results of the device's sensing
and fluid
detection. As use herein, a fluid is a unit of a vapor, liquid, solution, gas,
or other forms,
and mixtures thereof, of a substance being analyzed. Thus, a fluid sample can
include
chemical analytes, odors, vapors, and others. The sample can comprise a single
analyte
or a plurality of analytes.
A tubular wand 13 having an externally mounted sensor 14 and an
exhaust port 15 are provided to respectively receive and discharge samples to
be
analyzed. In certain embodiments, the externally mounted sensor is a plug-in
sensor
module. The operation of electronic circuitry of sensor modules, similar to
the externally
mounted sensor module of the present invention, is described in detail in U.S.
Patent No.
6,085,576, issued July 11, 2000, to Sunshine et al.
5
CA 02391451 2002-05-13
WO 01/36961 PCT/US00/31515
In one embodiment, the externally mounted sensor module of the present
invention incorporates "swap and sniff ' technology. Advantageously, the
externally
mounted modules can be easily swapped to compensate for various analytes or
for
specific environmental conditions. In certain aspects, a sensor module is
configured to be
releasable engageable into an external portion of a handheld device.
The sensor module includes a casing having at least two sensors and an
electrical connector. The casing is sized and configured to be received in an
external
receptacle of the sensing apparatus. The electrical connector is configured to
be
releasably engageable with a mating electrical connector of the sensing
apparatus when
the sensor module is received in the receptacle. The electrical connector
transmits the
characteristic signals from the sensors to the sensing apparatus. In certain
embodiments,
the characteristic sensor parameters and data are stored in a memory devices
such as an
electrically programmable ROM (EPROM), an electrically erasable and
programmable
PROM (EEPROM), and other memory technologies, integrated in the sensor module.
As
will be apparent to those of skill in the art, the externally mounted sensors
does not
preclude the presence of an internally mounted sensor.
In certain aspects, the sensor module contains a heating element. An on-
board processor can be used to provide temperature control for each individual
sensor
array device in the sensor module. In one implementation, each sensor array
device can
include a backside heater. Further, the processor can control the temperature
of the
sample chambers either by heating or cooling using a suitable thermoelectric
device.
Moreover, the analyte vapor can be heated and cooled within strict temperature
limits.
As illustrated in Figure 2 the response of a sensor array using the
foregoing movable or retractable sensor module embodiment is shown. At point
"A", the
sensor module is away from the sample in a first position. The sensor module
is then
placed in a second position, wherein the sample module is over the sample or
area to be
tested and the sensor responds (point "B"). Thereafter, the sensor is moved
back to the
first position. Using this configuration for a sensing device, it is possible
to take
advantage of the decrease in concentration as a function of distance to
calibrate or
reference the sensing device. With reference to Figure 1 for example, the
tubular wand
13 having an externally mounted sensor 14 can be telescopically retracted to
provide the
first position. In the fully retracted position the sensor module is in the
first position. In
the fully extended position the sensor module is in the second position.
6
CA 02391451 2002-05-13
WO 01/36961 PCT/US00/31515
B. Externally Mounted Sensor Module Having Two Sensor Elements
In another aspect, the present invention provides a sensing device for
detecting an analyte, comprising: a housing; a first sensor element
incorporating a first
array of sensors and a second sensor element incorporating a second array of
sensors
wherein both sensor elements are mounted externally on a housing. In certain
embodiments, the first sensing element is designed to sense a vapor, and is
referred to
herein as the sensing element. In certain aspects, the second sensor element
is designed
as a reference, and is referred to herein as the referencing element. In a
preferred aspect,
the first sensor element is a first array of sensors and the second sensor
element is a
second sensor array, wherein the first and second sensor arrays comprise
sensors that are
compositionally similar or the same. As will be apparent to those of skill in
the art, the
externally mounted sensors does not preclude the presence of an internally
mounted
sensor.
As shown in Figure 3, the artificial olfaction device 20 has a tubular wand
23 that has an externally mounted first sensor module 24 and an externally
mounted
second sensor module 25. Preferably, the first sensor module comprises a
sensor array
and the second sensor module comprises a sensor array having similar or the
same sensor
type. For example, if sensor element in 24 comprises surface acoustic wave
sensors, the
referencing sensor element in 25 will also comprise surface acoustic wave
sensors. In
this manner, both the first sensor element and the second sensor element have
similar or
the same sensor type.
In other aspects, the device comprises two sensor elements that are
externally mounted, wherein the second sensor element is positionally located
differently
(e.g., further away from the object to be measured) than the first sensor
element. In this
aspect, a vapor concentration gradient exists to provide a "reference" and a
real
measurement. The first sensor element closest to the sample provides the real
measurement. The second sensor element, i.e., the reference sensor element, is
located
further away from the object to be measured and thus provides a reference.
In a further aspect, the present invention provides a device having two
sensor elements wherein the location of the two sensors and differential vapor
detection
can be used for analyte determination. For instance, two sensors i.e., sensor
element 1
and sensor element 2, can be separated by some distance, "d," and if the vapor
at point 1
is different than at point 2 (point 1 and point 2 are separated by distance
"d") there will be
7
CA 02391451 2002-05-13
WO 01/36961 PCTIUSOO/31515
a differential response between sensor element 1 and sensor element 2. This
differential
response can be used to reference the two sensors with respect to drift.
In addition to referencing, the sensors can be used to advantageously map
a surface, such as a planar surface. For example, if the two sensors are
separated by 1
inch i.e., "d"= 1 inch, and one sensor is placed over the test sample and the
other sensor is
not over the test sample, this will create a differential signal. However, if
both sensor
elements are over a test sample or not over a test sample, the differential
reading will be
close to zero. This differential response is useful for mapping a surface. If
the objective
is to locate contamination on a surface, for example, a countertop, dirty
floorboards,
hazardous leaks, fecal matter, anywhere there is a point source having a
chemical profile,
the edge of the contamination will signal a positive response. This is
advantageous
because any deviation indicates a response. In general, a deviation from a
zero response
is typically more sensitive than trying to see a small change on a large
baseline.
As such, in certain embodiments, the present invention relates to mapping
an x-y surface for detection of an analyte, the method comprising: moving in
tandem at
least two sensor arrays separated by a distance "d" across an x-y surface to
produce a
plurality of responses; analyzing the responses and thereby mapping the x-y
surface for
detection of an analyte. In this embodiment, the two sensor arrays are
separated by a
distance "d". If one sensor array is placed over an analyte, the analyte being
on the x-y
surface, and the other sensor is not over the analyte, this will create a
differential signal
between the sensor arrays. However, if both sensor arrays are over the analyte
or,
alternatively, not over the analyte, the differential reading will be small or
close to zero.
The resolution of the analyte on the x-y surface is inversely proportional to
the distance
"d". The greater the resolution required, the smaller the distance separating
the tandem
sensor array.
Using the methods of the present invention, it is also possible to increase
the dynamic range of the sensor arrays. The dynamic range is characterized by
the lowest
detectable amount of analyte, which is given by the noise of the sensor
response, and the
maximum detectable amount of analyte that is given by the saturation effects
of the
sensor. Using the devices of the present invention, it is possible to reduce
the noise level
of the sensor system thereby increasing the dynamic range of the sensor
arrays.
8
CA 02391451 2002-05-13
WO 01/36961 PCT/US00/31515
C. Pneumatics
As shown in Figure 4, in certain embodiments, the present invention
relates to an increased duty-cycle pneumatic sensing train 30. In a preferred
embodiment,
the sensor module comprises at least two pneumatic vapor paths 33, 35 and at
least two
sensor arrays 36, 37. In a preferred differential flow pneumatic train 30 of
the present
invention, sensors 36, 37 can be independently externally mounted, internally
mounted,
or alternatively, one sensor can be externally mounted and one sensor can be
internally
mounted.
In operation, sensor 36 is used to detect an analyte. While sensor 36 is
being purged, sensor 37 can be used for detection or vise versa. Using the
differential
flow pneumatics of the present invention, it is possible to increase the
frequency of
analyte detection. In certain other embodiments, the pneumatics permits
switching
between a calibration source and an analyte source. Where sensor calibrations
are
frequent, the module of the present invention provides the ability to switch
gases on a
schedule consistent with desired pre-programmed calibration cycles.
In a further aspect, the present invention relates to sensor pneumatics
comprising a pump having a reverse flow feature. The pump functions in
alternative
modes wherein in the first mode, sample air is taken in, and in a reverse
mode,
background air purges the system. In this manner, the duty cycle can be
increased over
conventional pumps.
D. Pasivasion Laver
In other embodiments, the present invention provides a sensing device
comprising a first sensor element and a second sensor element that are
physically located
in spatially similar or identical positions with regard to the analyte;
however, the analyte
is prevented or blocked from contacting the second sensor element (i.e., the
reference
sensor). In this embodiment, a physical barrier exists between the reference
sensor
element and the analyte to be identified. Preferably, the sensing element is
pasivated with
a material to prevent the analyte from contacting the surface of the sensing
element.
Suitable pasivation materials include, but are not limited to, Si02 and Si02
based films,
thermal oxides, silane, SiH4, Si3N4, tetraethoxysilane, Si(OCzH5)4, boro
silicate glasses,
and spin on glass.
Figure 5 shows a top sectional view of an embodiment of a sensor module
that includes four plug-in sensor devices 41A and 41B within a single cavity
or sample
9
CA 02391451 2002-05-13
WO 01/36961 PCT/US00/31515
chamber 42. Disposed atop sensor array 41A is a pasivation layer 43 (hatched
lines) to
prevent or block the analyte from contacting sensor array 41A.
In another embodiment, the sensing device is configured so vapors do not
contact the surface of the second sensor element i.e. the reference element.
This can be
accomplished by the use of a second sensor module. The reference-sensing
module
completely encloses the reference sensors, thereby preventing vapors to
contact the
surface of the sensors. With regard to Figure 6, a top sectional view of an
embodiment
having two modules 51 and 54 wherein the first module is a sensing module 51
that
includes 2 plug-in sensor devices 52 and 52A. The second sensing element 54 is
a
referencing element and also includes 2-plug-in sensor devices 53 and 53A.
Sample
chamber 54 is defined, in part, by a cover 55 (Fig. 6B) that is secured over
module 54.
In another aspect, the reference element has a porous membrane associated
therewith. In this way, the analyte's contact with the sensor array is slowed.
The porous
membrane limits diffusion to the reference sensor. This process of limited
diffusion of
the analyte allows sampling of the sensors at different points of time and
thus, referencing
and calibration can be done simultaneously. The sensors are identical and thus
the
responses are identical. The pasivation material only slows diffusion and is
not analyte
selective. Suitable porous pasivation layers include, but are not limited to,
porous
plastics, Teflon, and dialysis materials.
Moreover, the pasivation layer can reduce or eliminate humidity. Using an
absorption or adsorption material especially designed for water vapor, the
pasivation layer
can reduce or eliminate humidity in the test sample.
E. Methods to Reduce Drift
Instabilities and drift are serious problems in chemical sensors and could
effect the identification of analytes. Noncumulative drift denotes statistical
variations of
the sensor signal or response. Cumulative drift leads to irreversible changes
of calibration
and can result from sensor deterioration. Figure 7 is an xy plot 70 of sensor
signal versus
time. Short-term drift can occur after switching on the sensor array device
71. These
short-term drifts are caused by the time required to establish steady-state
conditions, such
as a constant operational temperature of the sensor array. In certain
instances, thermal
drift is related to changes of the sensor signal upon variations of ambient
temperature.
Sensor signals 74 and 75 show actual analyte sensing. Thermal drift can be
reduced or
CA 02391451 2002-05-13
WO 01/36961 PCT/US00/31515
eliminated by maintaining the sensor module at a uniform temperature. It is
possible to
reduce, compensate or eliminate drift using differential temperature
measurements.
As such, the present invention provides a method for reducing drift by
using differential thermal measurements. Thus, in another embodiment, the
present
invention provides a method for reducing drift in an array of sensors,
comprising:
contacting the array of sensor with an analyte at a first temperature to
produce a first
response; contacting the array of sensor with the analyte at a second
temperature to
produce a second response; and subtracting the first response from the second
response
thereby reducing drift.
In this embodiment, the need to take a background response requirement
for a baseline has been alleviated. Prior to the present invention, analyte
detection
required the background or ambient response to be taken as a reference. This
background
or reference sample is then subtracted from the test response. This method can
be
inefficient because the background has to be purged before an analyte can be
measure.
To increase the sensor array duty cycle, it has been discovered that the array
of sensors
can measure the analyte at two temperatures, and thereby alleviate the purge
cycle. This
dramatically increases sensor sampling and duty cycle. In certain aspects, the
sampling is
performed at two temperature values, wherein the temperature values differ
between
about 5 C and about 150 C. More preferably, the temperatures differ between
about 2 C
to about 70 C. In certain aspects, the analyte and the sensor array are
equilibrated at the
first temperature. In addition, the analyte and the sensor array are
optionally equilibrated
at the second temperature. In certain embodiments, the artificial olfaction
device
comprises two arrays of sensors.
In another aspect of differential measurements, the present invention
relates to a method for reducing drift by using differential sensor
measurements. Similar
to thermal differences, the use of sensor arrays having various sensor
thickness results in
eliminating drift. Thus, in yet another embodiment, the present invention
provides a
method for reducing drift in an array of sensors, comprising: contacting a
first sensor
having a first sensor thickness with an analyte to produce a first response;
contacting a
second sensor having a second sensor having a second sensor thickness with the
analyte
at a second temperature to produce a second response; and subtracting the
first response
from the second response thereby reducing drift.
11
CA 02391451 2008-03-31
F. Methods to Calibrate
In general, for unequivocal characterization of their dynamic and static
properties, sensors have to be calibrated and tested. In the simplest case,
the calibration
curve for sensor arrays is linear. The devices of the present invention
provide internal
diagnostics and built-in self-calibration features, which allow for improved
performance.
Moreover, the sensors of the present invention have the ability to perform
internal
diagnostics and self-calibration, thereby validating that the sensor is
operating within
acceptable tolerances. In certain embodiments, the devices and methods of the
present
invention provide the means to automatically calibrate in-situ sensor arrays,
for many
different analyte mixtures. The sensor calibration is routinely scheduled over
an
extended period at the user's discretion. In certain aspects, the devices of
the present
invention provide an interface to record, display and analyze sensor data in
real time
against an analyte standard.
II. SENSOR ARRAYS
The devices and methods of the present invention include an array of
sensors and, in certain instances, the sensors as described in U.S. Patent No.
5,571,401
are used. Sensors suitable for detection of analytes associated with
agricultural products
include, but are not limited to, surface acoustic wave (SAW) sensors; quartz
microbalance sensors; conductive composites; chemiresitors; metal oxide gas
sensors,
such as tin oxide gas sensors; organic gas sensors; metal oxide field effect
transistor
(MOSFET); piezoelectric devices; optical sensors; sintered metal oxide
sensors; Pd-gate
MOSFET; metal FET structures; conducting-and nonconducting regions-disposed on
metal FET structures; metal oxide sensors, such as a Tuguchi gas sensors;
phthalocyanine sensors; electrochemical cells; conducting polymer sensors;
catalytic gas
sensors; organic semiconducting gas sensors; solid electrolyte gas sensors;
temperature
sensors, humidity sensors, piezoelectric quartz crystal sensors; and Langmuir-
Blodgett
film sensors.
In a preferred embodiment, the sensors of the present invention are
disclosed in U.S. Patent No. 5,571,401. Briefly, the sensors described therein
are
conducting materials and nonconducting materials arranged in a matrix of
conducting
and nonconducting regions. The nonconductive material can be a nonconducting
12
CA 02391451 2008-03-31
polymer such as polystyrene. The conductive material can be a conducting
polymer,
carbon black, an inorganic conductor and the like. The sensor arrays comprise
at least
two sensors, typically about 32 sensors and in certain instances 1000 or more
sensors.
The array of sensors can be formed on an integrated circuit using
semiconductor
technology methods, an example of which is disclosed in PCT Publication WO
99/08105, entitled "Techniques and Systems for Analyte Detection," published
February
19, 1999. Another preferred sensor is disclosed in WO 99/27357 entitled
"Materials,
Method and Apparatus for Detection and Monitoring Chemical Species," published
June
3, 1999.
In one embodiment, the sensor arrays are formed from composites of
poly(3,4-ethylenedioxy)thiophene-poly(styrenesulfonate) as a conductive
component
with an insulating polymer (see, Solzing et al., Anal. Chem., 72, 3181-3190
(2000)). The
insulating polymers can be for example, poly(vinylacetate),
poly(epichlorohydrin),
poly(ethylene oxide), etc.
In other instances, the sensors are disclosed in WO 00/00808, published
on January 6, 2000, to Lewis et al. Chemical sensors are disclosed comprising
a
plurality of alternating nonconductive regions (comprising a nonconductive
material)
and conductive regions (comprising a conductive material), wherein the
conducting
region comprises a nanoparticle.
Preferably, the sensor arrays of the present invention comprise at least
one sensor selected from the following group of sensors, inorganic metal oxide
semiconductors such as tin-oxide based sensors, intrinsically conducting
polymers such
as polymers of pyrrole, thiophene and aniline, mass sensitive piezoelectric
sensors such
as bulk acoustic wave and surface acoustic wave sensors, polymer compositions
on
metal FET, and nonconducting/conducting regions sensors.
As will be apparent to those of skill in the art, the sensors making up the
array of the present invention can be made up of various sensor types as set
forth above.
For instance, the sensor array can comprise a conducting and nonconducting
regions
sensor, a SAW sensor, a metal oxide gas sensor, a conducting polymer sensor, a
Langmuir-Blodgett film sensors, polymer composites on metal FET, and
combinations
thereof.
13
CA 02391451 2008-03-31
In certain embodiments, the temporal response of each sensor (response
as a function of time) is recorded and can be displayed. Various responses
include, but
are not limited to, resistance, impedance, capacitance, inductance, magnetic,
work
function, optical, etc. The temporal response of each sensor can be normalized
to a
maximum percent increase and percent decrease that produces a response pattern
associated with the exposure of the analyte. By iterative profiling of known
analytes, a
structure-function database correlating analytes and response profiles is
generated.
Unknown analytes can then be characterized or identified using response
pattern
comparison and recognition algorithms. Accordingly, analyte detection systems
comprising sensor arrays, a measuring device for detecting responses across
each sensor,
a computer, a display, a data structure of sensor array response profiles, and
a
comparison algorithm(s) or comparison tables are provided. In another
embodiment, the
electrical measuring device or detector is an integrated circuit comprising
neural
network-based hardware and a digital-analog converter (DAC) multiplexed to
each
sensor, or a plurality of DACs, each connected to different sensor(s).
In certain embodiments, the present invention provides an array of an
array of sensors. As used herein, an array of an array of sensor is termed a
massively
parallel independent array (MPIA). This device is a matrix of sensors that can
sense
multiple bottles or vessels simultaneously. The device is especially useful
for assaying
or for diagnostic purposes for multiple vessels. For example, in a
combinatorial library
of catalysts, a MPIA can be used to simultaneously determine catalysts having
unique
signature patterns of interest. Using the MPIA systems of the present
invention it is
possible to monitor the efficiency of antibiotics, catalysts, drugs,
biomolecule binding
efficiencies, nucleic acid hybridizations, ligand-ligand interactions,
biomolecule
interactions, potential drug candidates, etc. See, WO 99/53300, published
April 13,
1999, to Lewis et al. WO 99/53300 discloses chemical sensors for detecting the
activity
of a molecule or analyte of interest. The chemical sensors comprise an array
or plurality
of chemically sensitive resistors that are capable of interacting with the
molecule of
interest, wherein the interaction provides a resistance fingerprint. The
fingerprint can be
associated with a library of similar molecules of interest to determine the
molecule's
activity.
14
CA 02391451 2008-03-31
In one embodiment, the MPIAs of the present invention are fabricated
using combinatorial techniques. Devices for the preparation of combinatorial
libraries
are commercially available (see, e g., 357 MPS, 390 MPS, Advanced Chem Tech,
Louisville KY, Symphony, Rainin, Wobum, MA, 433A Applied Biosystems, Foster
City, CA, 9050 Plus, Millipore, Bedford, MA). A number of well known robotic
systems have also been developed for solution phase chemistries. These systems
include
automated workstations like the automated synthesis apparatus developed by
Takeda
Chemical Industries, LTD. (Osaka, Japan) and many robotic systems utilizing
robotic
arms (Zymate II, Zymark Corporation, Hopkinton, Mass.; Orca, Hewlett-Packard,
Palo
Alto, Calif.) which mimic the manual synthetic operations performed by a
person of
skill in the art. The nature and implementation of modifications to these
devices (if any)
so that they can operate as discussed herein will be apparent to skilled
artisans.
Moreover, combinatorial methods of making sensors are disclosed in WO
99/00663,
published January 7, 1999, to Lewis et al. The methods disclosed therein
combine
various ratios of at least first and second organic materials to fabricate
sensors.
The MPIA can utilize sensors as disclosed in WO 99/40423, published
August 12, 1999, to Lewis et al. Arrays of sensors useful for analyzing chiral
analytes
and producing a sample output are disclosed. The array comprises
compositionally
different sensors, wherein a sensor comprises a chiral region. The analyte
generates a
differential electrical response across the sensor thereby being detected.
In certain instances, the sensor arrays comprise sensors having aligned
particle based sensor elements as disclosed in WO 00/33062, published June 8,
2000, to
Sunshine et al. The sensor arrays disclosed therein comprise first and second
sensors
wherein the first sensor comprises a region of aligned conductive material;
electrically
connected to an electrical measuring apparatus. The aligned conductive
material
improves the signal to noise of vapor sensors allowing lower detection limits.
Such
lower detection limits allow for the identification of lower concentrations of
hazardous
material and is advantageous in medical applications, such as the detection of
disease
states.
In certain other embodiments, the sensor arrays are chemically sensitive
resistors wherein the resistors are composed of a conductor (e.g. carbon
black) and a
conducting polymer, such as polyaniline. The polyaniline composites can be
used to
CA 02391451 2008-03-31
detect biogenic amine odorants such as putrescine, cadaverine and spermine
(see,
Sotzing et aL, Chem. Mater. 12, 593-595 (2000)).
III. PRECONCENTRATOR
In certain aspects, the present invention optionally comprises a
preconcentrator. In this aspect, a volume of the gas to be sampled is
introduced into a
sample chamber where it is transported by means of convention, such as
convection,
into the vicinity of the sorbent material. Suitable transporting means
include, but are not
limited to, a fan, an air pump, or it can be means for heating the cylindrical
container to
15a
CA 02391451 2002-05-13
WO 01/36961 PCT/US00/31515
create a convective air flow between the inlet and the outlet. The sorbent
material is
chosen from known materials designed for the purpose of sorbing gases, vapors,
and the
like. In certain embodiments, the sorbent material includes, but is not
limited to, a
nanoporous material, a microporous material, a chemically reactive material, a
nonporous
material and combinations thereof. Such absorbents include, for example,
activated
carbon, silica gel, activated alumina, molecular sieve carbon, molecular sieve
zeolites,
silicalite, A1PO4, a polymer, a co-polymer, polymer blends, alumina and
mixtures thereof.
In certain embodiments, the absorbent has a pore size from about 1 nm to about
100 nm
and, preferably, from about 1 nm to about 50 nm.
Suitable commercially available adsorbent materials are disclosed in U.S.
Patent No. 6,085,576 and include, but are not limited to, Tenax TA, Tenax GR,
Carbotrap, Carbopack B and C, Carbotrap C, Carboxen, Carbosieve SIII, Porapak,
Spherocarb, and combinations thereof. Preferred adsorbent combinations
include, but are
not limited to, Tenax GR and Carbopack B; Carbopack B and Carbosieve SIII; and
Carbopack C and Carbopack B and Carbosieve SIII or Carboxen 1000. Those
skilled in
the art will know of other suitable absorbent materials.
After sometime period that is chosen to be adequate for sorbing the desired
analytes from the vapor phase onto the material, the circulation is stopped
and then the
material is desorbed from the sorbent phase and released into the sensor
chamber. The
desorbing of the concentrated analyte from the sorbent can be accomplished by
thermal
means, mechanical means or a combination thereof. Desorption methods include,
but are
not limited to, heating, purging, stripping, pressuring or a combination
thereof.
In certain embodiments, the sample concentrator is wrapped with a wire
through which current can be applied to heat and thus, desorb the concentrated
analyte.
The analyte is thereafter transferred to the sensor array.
The process of sorbing the material onto the sorbent phase not only can be
used to concentrate the material, but also can be advantageously used to
remove water
vapor. The water vapor is preferably removed prior to concentrating the
analyte;
however, in various embodiments, the vapor can be removed concomitantly or
after the
analyte is concentrated. In a preferred embodiment, the water vapor is removed
prior to
presenting the desired analyte gas mixture to the sensor array. Thus, in
certain
embodiments, the fluid concentrator contains additional absorbent material to
not only
concentrate the analyte, but to remove unwanted materials such gas
contaminates and
moisture.
16
CA 02391451 2008-03-31
IV. ALGORITHMS
The device and methods of the present invention optionally comprise
pattern recognition algorithms. Many of the algorithms are neural network
based
algorithms. A neural network has an input layer, processing layers and an
output layer.
The information in a neural network is distributed throughout the processing
layers. The
processing layers are made up of nodes that simulate the neurons by its
interconnection
to their nodes.
In operation, when a ANN is combined with a sensor array, the sensor data is
propagated through the networks. In this way, a series of vector matrix
multiplications
are performed and unknown analytes can be readily identified and determined.
The
neural network is trained by correcting the false or undesired outputs from a
given input.
Similar to statistical analysis revealing underlying patterns in a collection
of data, neural
networks locate consistent patterns in a collection of data, based on
predetermined
criteria.
Suitable pattern recognition algorithms include, but are not limited to,
principal component analysis (PCA), Fisher linear discriminant analysis
(FLDA), soft
independent modeling of class analogy (SIMCA), K-nearest neighbors (KNN),
neural
networks, genetic algorithms, fuzzy logic, and other pattern recognition
algorithms. In a
preferred embodiment, the Fisher linear discriminant analysis (FLDA) and
canonical
discriminant analysis (CDA) and combinations thereof are used to assess
patterns in
responses from the electronic noses of the present invention. Operating
principles of
various algorithms suitable for use in the present invention have been
disclosed (see,
Shaffer et al., Analytica Chimica Acta, 384, 305-317 (1999)). The Fisher
linear
discriminant analysis as it pertains to artificial olfaction is disclosed in
WO 99/61902,
published December 2, 1999, to Lewis et al.
V. NETWORKED SYSTEMS
In certain instances, the devices, methods and apparatus of the present
invention can be used in a networked environment. For instance, the networked
systems
of the present invention allow the methods to be carried out in one location
such as with
a handheld device and subsequently transmit digital signals over a computer
network,
such as the Internet, for analysis at a remote location. Suitable methods and
systems for
17
CA 02391451 2008-03-31
detecting and transmitting sensory data over a computer networked are
disclosed in WO
00/52444, published September 8, 2000, to Sunshine et al.
As disclosed therein, communication between the on-board processor of
an artificial olfaction device and the host computer is available to configure
the device
and to download data from or to the outside world, in real time or at a later
time via a
number of communication interfaces including, but not limited to, an RS-232
interface,
a parallel port, an universal serial bus (USB), an infrared data link, an
optical interface
and an RF interface. Serial communications to the outside world are provided
by the on-
board low power RS-232 serial driver. Communication to the outside world
includes,
but is not limited to, a network, such as a computer network e.g. the Internet
accessible
via Ethernet, a wireless Ethernet, a token ring, a modem, etc. A transfer rate
of 9600
bits/second can transmit approximately 400 data points/second, and higher
transfer rates
can be used.
The computer network can be one of a variety of networks including a
worldwide computer network, an Internet, the Internet, a WAN, a wireless
network, a
LAN or an intranet. It should be understood that conventional access to the
computer
network is conducted through a gateway. A gateway is a machine, for example, a
computer that has a communication address recognizable by the computer
network.
VI. EXAMPLES
The following examples are offered to illustrate, but not to limit the
claimed invention.
Example 1
This Example illustrates an e-nose device having two sensor arrays
wherein sensor element 2 has a porous membrane associated therewith.
In this Example, the detection and identification of analytes will be
accomplished by using an electronic nose having two 32-sensor arrays. Sensor
element 1
(having 32-sensors) is a sensing array and sensor element 2 (having 32-
sensors) is a
referencing array. Sensor element 2 has a porous membrane associated
therewith. The
analyte's contact with the reference sensor array will thus be slowed. The
porous
membrane limits diffusion to the 2nd sensor. This process of limited diffusion
of the
analyte allows sampling of the sensors at different points of time and thus,
referencing
and calibration can be done simultaneously.
18
CA 02391451 2002-05-13
WO 01/36961 PCT/US00/31515
A Keithley electrometer and scanner will be used to scan the resistances of
two 32-sensor arrays during the experiment. In certain instances, the
temperature of the
substrates will not be controlled and the measurements will be done at room
temperature.
For each sample test, there will be 60 seconds of background recording (purged
with air),
120 seconds of exposure time, 120 seconds of recovery time (purged with air
with RH
level of about 3%), 180 seconds of recovery without recording the data (purged
with air),
and 30 seconds of final recording time (purged with air).
The response patterns from the two 32-sensor array will have good
reproducibility. The response (the normalized resistance change, (Rmax-
Ro)/Ro), where
Rmax and Ro are the maximum and base (initial) resistance, respectively) of
each of the
sensor arrays to each sample tested will be employed to form a covariance
matrix, which
is used to do principal component analysis. PCA of the analytes plus control
will be
clearly discriminated by the sensor array. SIMCA is also used to evaluate the
data.
As shown in Figure 8, the sensor response 81 at time t, where the first
sensor is just beginning to respond 82 will give a AR/R value and the second
sensor array
will give a second response 83. The two individual AR/R values (for the first
and second
trace) can be used to calibrate the system. The sensor arrays are identical
and therefore,
the two responses are identical. The pasivation material on the second sensor
only slows
diffusion and is not selective.
Example 2
This Example illustrates differential temperature measurement of sensor
arrays.
In this method an analyte is detected at two temperatures. The first
sampling is conducted at ambient temperature and the second temperature is at
60 C. In
this experiment, the drift of the sensor array can be reduced using
differential thermal
measurements. Thus, by contacting the array of sensor with an analyte at a
first
temperature to produce a first response and subsequently contacting the array
of sensors
with the analyte at a second temperature to produce a second response and
thereafter,
subtracting the first response from the second response the drift can be
reduced. Thus, the
need to take a background response has been alleviated. Using this method,
dramatic
increases in sensor sampling and duty cycle are achieved.
19
CA 02391451 2002-05-13
WO 01/36961 PCT/US00/31515
Example 3
This Example illustrates the use of a massively parallel independent array
(MPIA) to monitor a reaction such as the conversion of one reactant to another
using a
combinatorial library of catalysts. The method is a way of evaluating catalyst
activity.
A different catalyst is loaded into small wells of a microtiter plate (e.g.,
96-well or 384-well microtiter plates) together with the target reactants. In
certain
aspects, the catalysts are prepared using combinatorial techniques. Such
catalysts include
for instance, palladium on carbon (having various weight percents of palladium
e.g., 1%,
2%, 3%, etc.), Raney nickel, Raney copper, etc. The MPIA is mounted on the
headspace
of the microtiter plate for real time monitoring of the conversion process
related to
catalytic activity. In a preferred embodiment, above each well in the
microtiter plate, is a
sensor array (i.e., at least two sensors). Thus, in certain aspects, such as
in a 96 well
format, there are at least 192 total sensors in the MPIA. In a 384 format,
there are in
certain aspects, at least 768 sensors in the MPIA.
Specific examples of catalytic activity include, but are not limited to, the
hydrogenation of 1-hexyne to 1-hexene or to hexane (i.e. the fully saturated
hydrocarbon)
using a variety of oxides as the catalysts. The decrease in 1-hexyne
concentration and the
increase in concentration of the saturated hydrocarbons can be monitored using
the sensor
array(s) and analyzed with a computer in real time. Complete conversion of the
1-hexyne
can also be determined.
Another specific example includes the dehydrogenation of cyclohexane to
benzene using a library of solid-state catalysts. The decrease in cyclohexane
concentration and the increase in benzene concentration can be monitored using
a sensor
array(s) and the conversion monitored real time.
Advantageously, independent response patterns for each sensor array are
simultaneously monitored and compared. Thus, using the MPIA in a 96-well
format of
the present invention, sensor array above well 1-96 is compared with sensor
array 33-96
and so forth. Thus, each sensor array within the MPIA (1-96, 2-96, 3-96, 4-96,
etc.) is
simultaneously and independently monitored. In operation, a matrix of response
patterns
is generated and compared using pattern recognition algorithms. In certain
aspects, the
efficiency of the reactions is monitored in real time. Preferably, the MPIA
system resides
in a networked environment. Using the MPIA systems of the present invention it
is
possible to monitor the efficiency of antibiotics, catalysts, drugs,
biomolecule binding
CA 02391451 2008-03-31
efficiencies, nucleic acid hybridizations, ligand-ligand interactions,
biomolecule
interactions, drug candidates, etc.
It is understood that the examples and embodiments described herein are
for illustrative purposes only and that various modifications or changes in
light thereof
will be suggested to persons skilled in the art and are to be included within
the spirit and
purview of this application and scope of the appended claims.
21
