Note: Descriptions are shown in the official language in which they were submitted.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-I-
SYSTEMS FOR DIFFERENTIAL ION MOBILITY ANALYSIS
5.
I~EELof~TED APPLICATI~T~TS
'T°°hgs applicati~x~ gs a c~antinuation-ire-pa~c ~f U.S.
Application TAT~. IO1697~708~ filed ~~a
Oct~beg 30, 2003, v~hich claina~ tlae benefit ~f fr~visi~nal Application 1~~.
60/4~!2,53~~, filed
October 31, 2002.
'I'his application is a contineaation-in-part of "Systems for I~gfferent~al
I~n 1~A~bghty
Analysis", as U.S. Application T~To. 10/794,776, filed larch 5, 20049 (Attoa-
ney l~ocl~et
TlTun~bex 3239.1051-00), which claims the benefit of Provisional Application
hTo. 60/453,4.89
filed March 10, 2003.
This application claims the benefit of U.S. Provisional Application.No.
60/453,451,
IS filed on March 10, 2003 and U.S. Provisional Application No. 60/530,815,
filed on
December 18, 2003.
This application,is a continuation-in-part of US Application 10/462,206, filed
June 13,
2003, which is a continuation-in-part of U.S. Patent Application No.
10/321,822 filed
December 16, 2002, a continuation-in-part of U.S. Patent Application No.
10/123,030 filed
April 12, 2002, and a continuation-in-part of U.S. Patent Application No.
10/187,464 filed
June 28, 2002, and claims the benefit of U.S. Provisional Application No.
60/389,400 filed
June 15, 2002, claims the benefit of U.S. Provisional Application No.
601398,616 filed July
25, 2002, claims the benefit of U.S. Provisional Application No. 60/418,671
filed October I 5,
2002, claims the benefit of U.S. Provisional Application 60/453,287 filed
March 10, 2003,
2S and claims the benefit of U.S. Provisional Application No. 601468,306 filed
May 6, 2003.
This application claims the benefit of Provisional Application No. 60/549,004,
filed
March l, 2004, entitled. "Chemical Agent Detector" by Raanan A. Miller et al.,
(Docket No.
M143; US Express Mail Label No. EO 902 877 387US.
SUBSTITUTE SHEET (RULE 26)
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-2-
This application is a continuation-in-part of US Application 10/321,822,
filed December 16, 2002, which is a continuation of US Application No.
09/358,312, filed July 21, 1999 (Patent No. 6,495,823).
This application is related to US l~pplication No. 10/18~,4~64., filed Juzle
28,
2002, which is a continuation-in-part of [Attorney Docket M007] U.S.
Applmation
No. 09/896,536 filed June 30, 2001 entitled "Apparatus For Simultaneous
Identification Of Multiple Chemical Compounds;" and claims the benefit of
[Attorney Doclcet M012] U.S. Provisional Application No. 60/340,894 filed
October
30, 2001 entitled "Compound Identification By Mobility Dependence On Electric
Field," [Attorney DocketM033] U.S. Provisional Application No. 60/334,670,
filed
November 15, 2001 entitled "System For Ion Mobility And Polarity
Discrimination
And Identification Of Chemical Compounds;" [Attorney Docket M033R]U.S.
Provisional Application No. 60/340,904, f led December 12, 2001 entitled
"System
For Ion MobilityAnd Polarity Discrimination And Identification Of Chemical
Compounds;" [Attorney Docket M041] U.S. Provisional Application No.
60/342,588 filed December 20, 2001 entitled "Field Dependence Of Mobilities
For
Gas Phase Protonated Monomers And Proton Bound Dimers Of Ketones By Planar
Field AsymmetricWaveform Ion Mobility Spectrometer (PFAIMS);" and [Attorney
Docket M042] U.S. Provisional Application No. 60/351,043 filed January 23,
2002
entitled "Method And Apparatus For FAIMS Detection Of SF6".
The entire teachings of these disclosures are incorporated herein by
reference.
BACKGROUND OF THE INVENTION
Spectrometers are used in chemical analysis for identification of compounds
in a sample. In some cases a quick indication of presence of particular
compounds in
a sample is needed, while at other times the goal is complete identification
of all
compounds in a chemical mixture. Accordingly, samples array be taken directly
from
the environment and analysed or may be prepared by processing andlor
separating
the constituents before spectrometric analysis.
Spectrometers based on ion mobility have been used to detect various
chemical and biological compounds. Such spectrometers include ion-mobility
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-3-
spectrometry (IMS) and differential ion mobility spectrometers (DMS) which are
also l~nown as field asymmetric wavefonn ion mobility spectrometers (FAIMS)
Commercially available IMS systems are based on time-of flight (TOF-
1MS), L.~'., they nlea8llr~ the tlnle It t~eS lon5 t~ travel fT~111 a Shutter-
gate tS~ a
detector through an inert atmosphere (1 to 760 Torr.). The drift time is
dependent on
the mobility of ions in a low electric field based on size, mass and charge,
and is
characteristic of the ion species detected. TOF-IMS has been used for
detection of
many compounds including narcotics, explosives, and chemical warfare agents,
and
at least one TOF-IMS system has been adapted for use in a field-portable
device for
detection of bacterial spores in the environment.
DMS devices offer an alternative to the low field TOF-IMS ion mobility
process. In DMS, ion filtering is achieved based on accentuating differences
in
mobility of ionized molecules in a high field. The high field mobility
differences are
used for "signature" identification of chemical species in an iouzed sample.
DMS
filtering is an efficient process, combining controlled neutralization of
unselected
ion species while passing selected ion species for detection.
There is a strong and continuing interest in improved approaches to sample
characterization, particularly as may be provided in compact and portable
devices.
SUMMARY OF THE 1NVENTTON
Practices of the present invention are directed to methods and devices for
detection and identification of analytes in samples using the mobility-based
signature that is produced when a volatilized sample is passed through a
differential
ion mobility spectrometry (DMS) device. Any volatized or volatilizable sample
can
be analyzed including organic, chemical, agricultural or biological samples.
In one
embodiment, the present invention includes using DMS to generate separation
data
and at least one other processing step that yields its own separation data.
This
additional separation step may be before or after DMS filtering. Analytes are
reliably identified based on this combination of data.
In one embodiment, the samples subjected to the analysis by the methods
and devices of the present invention are either normally existent in the
volatile state
or require volatilization. Analytes in a sample can be volatilized with or
without
fragmentation. Analyte volatilization and fragmentation can be achieved by any
of
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-4-
the techniques known in the art including pyrolysis, thermodesorption, laser
ionization, microwave heating or chemical transformation. Either prior to or
following the volatilization, analytes in a sample can be additionally
separated using
any of the techniques known in the art such as gas chromato~aphy.
Each analyte is detected by its ion-mobility based signature. This signature
is
expressed as stored spectrometric data uniquely identifying the species being
analyzed. The combination and relative abundances of various analytes in a
sample
forms a pattern that can be used to identify the entire sample by use of the
stored
reference data. Preferably the ion-mobility based signature is based on the
differential mobility of that species as experienced in the compensated I~MS
filter
field.
Analysis of physiological samples can identify diseases, monitor patient's
condition or provide forensic information; analysis of environmental samples
can
detect chemical or biological contamination, including agents of chemical and
biological warfare, or determine geochemical composition of soil and
sediments;
analysis of food quality samples can detect bacterial and chemical
contamination as
well as early signs of decomposition; analysis of chemical samples can be used
to
monitor small and industrial scale processes as well as safety conditions;
analysis of
biological samples can be used to identify microorganisms in pure or mixed
cultures
as well as assess efficiency of medication or other antibiotic compounds;
analysis of
industrial samples can be used to monitor the quality of the material;
analysis of
aginicultural samples can detect pesticides, herbicides as well as analyze
soil and
determine quality of crops.
Accordingly, one embodiment of the invention is directed to a method of
detection and identification of analytes in a sample by an ion mobility based
device,
comprising (a) obtaining a volatilized sample comprising markers that are
detectable
by an aspect of ion mobility (preferably by I?MS) ; and (b) directing at least
a
portion of the volatilized sample to a I~TvIS detection device to obtain a
mobility-
based signature of at least one marker, wherein the mobility-based signature
correlates with an analyte in the sample, thereby detecting and identifying at
least
one analyte in the sample.
In another embodiment, the present invention is a method of detection and
identification of analytes in a sample, comprising (a) volatilizing at least a
portion of
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-5-
the sample to produce a volatilized sample that includes markers detectable by
an
aspect of ion mobility; and (b) directing at least a portion of the
volatilized sample to
a DMS device, to obtain a mobility-based signature of at least one marker,
wherein
the mobility-based signature correlates with an analyte in the sample thereby
detecting and identifying at least one analyte in the sample.
In another embodiment, the present invention is a device for analysis of
samples (e.g., biological, chemical, organic, agricultural) using an aspect of
ion
mobility, comprising (a) a volatilization part; and (b) a differential ion
mobility
spectrometry (DMS) device connected to said volatilization part.
hl another embodiment, the present invention is directed to a field
asymunetric ion mobility detection system, comprising an input part and an
output
part, the input part including a volatilization part; at least a pair of
spaced insulated
substrates cooperating to define between them an enclosed flow path for the
flow of
ions from the input part to the output part; at least two electrodes opposite
each other
and defined in the flow path, the at least two electrodes including at least
one filter
electrode associated with each substrate to form an ion filter section; and an
electronics part configured to apply controlling signals to the electrodes,
and the
electronics part applying an asymmetric periodic signal across the filter
electrodes
for filtering the flow of ions in the flow path, the filter being compensated
to pass
desired ion species out of the filter section.
In one embodiment, the present invention is a method of detection and
identification of analytes in a sample by an ion mobility-based device,
comprising
directing a portion of a sample into a first separation device thereby
obtaining a first
profile; directing a portion of a sample into a second separation device
thereby
obtaining a second profile, wherein at least one of the first and the second
separation
devices is a DMS device; and (c) combining the first and the second profiles
thereby
identifying at least one analyte in a sample.
The instant invention advantageously employs differential mobility
spectrometry in a number of heretofore undisclosed industrial, clinical,
diagnostic
and environmental applications. The methods and devices of the present
invention
enable rapid detection and identification of compounds. Such detection and
identification can be made rapidly and with a high level of confidence.
Practices of
the invention are sensitive to parts per billion and even parts per trillion
levels.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-6-
Furthermore, unlike the devices of prior art, embodiments of the invention can
simultaneously filter and detect both positive and negative ions of an ion
species.
Systems of the invention may be used alone or in combination with other
analytical
equipment with increased likelihood of accurate identification of chemical
compounds, even at trace levels, and even for complex mixtures that heretofore
have
been difficult to resolve. As a result, an inexpensive, fast and accurate
chemical
marker (including biomarker) analysis system which can even be provided in a
compact and field-portable package.
BRIEF DESCRIPTION OF THE I~RAW1NGS
The patent or application file contains at least one drawing executed in
color.
Copies of this patent or patent application publication with color drawings
will be
provided by the Office upon request and payment of the necessary fee, as
needed.
The foregoing and other objects, features and advantages of the invention
1 S will be apparent from the following more particular description of
preferred
embodiments of the invention, as illustrated in the accompanying drawings in
which
like reference characters refer to the same parts throughout the different
views. The
drawings are not necessarily to scale, emphasis instead being placed upon
illustrating the principles of the invention.
FIG. 1 is a schematic flow diagram of an embodiment of the invention.
FIG. 2A is a schematic diagram of a separation system of the invention.
FIG. 2B is a schematic diagram of an analysis system of the present
invention.
FIG. 2C shows a detection scan according to the invention.
FIG. 3 is a schematic diagram of a separation system of FIG. 2A that
includes an SPME pre-separator.
FIG. 4~A is a schematic diagram of an embodiment of the present invention
with a cylindrical arrangement of the electrodes of a I~MS system.
FIG. 4B shows one embodiment ofI)MS electrodes where the electrodes are
cm~ed or curvilinear.
FIG. 5 is a schematic of au embodunent of the invention with a pyrolysis
front-end.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
_7_
FIG. 6 is a schematic diagram of an alternative separation method of the
present invention.
FIG. 7 shows a mass-spectrometric analysis of pyrolysis products of the
spores ofB. sa~btilis.
FIG. 8 shows positive and negative ion spectra for picolinic acid in practice
of the invention.
FIG. 9 shows positive and negative ion spectra for dipicolinic in practice of
the invention.
FIG. 10 shows positive and negative ion spectra for pyridine in practice of
the invention.
FIG. 1 lA shows the full time-dependent DMS spectrum of pyrolyzed B.
subtilis spores as a simulant for B. antlar~acis.
FIG. 11B shows the individual positive and negative ion spectra at 10
seconds after pyrolysis.
FIG. 12 shows spectra with putrescine and cadaverine resolved from one
another in practice of the invention.
FIG. 13 shows baclcground spectra with no sample present on a SPME fiber.
FIG. 14 shows spectra obtained from subject #1.
FIG. 15 shows spectra obtained from subject #2.
FIG. 16 shows spectra generated for markers for bacillus spore pyrolysis in
practice of the invention.
FIG. 17 shows positive ion spectra for urine headspace detected in practice
of the invention.
FIG. 18 shows spectra for a DMS embodiment of the invention with a GC
front-end.
FIG. 19 shows spectra for the GC-DMS where the chromatographic runtime
has been decreased leading to co-eluting species and showing that a practice
of the
invention is able to resolve the co-eluted species.
FIG. 20 shows comparison of prior art F~ and a DMS embodiment of the
invention for reproducibility for a homologous alcohol anixture.
FIG. 21 shows the results from py-GC-DMS characterization of positive ions
for E. coli (A), M. luteus (B), and B. megateriuyn (C).
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
_g_
FIG. 22 shows the results from py-GC-DMS characterization of negative
ions for E. coli (A), M. luteus (B), and B. megczte~~iufya (C).
FIG. 23 the profiles from py-GC-DMS analyses of E. c~li, Lipid ~ and
mixtures of Lipid ~ axld E. c~li.
FIG. 24 shows the compensation voltage versus retention time for the 50
peaks of highest intensity in py-GC DMS analyses of E. c~li (grey signals) and
1~1
luteus (black siglzals) for positive polarity ions.
FIG. 25 shows the compensation voltage versus retention time for the 30
peaks of highest intensity in py-GC-DMS analyses of E. c~li (grey signals) and
1VI.
luteus (black signals) for negative polarity ions.
FIG. 26 shows the effect of separation voltage on py-GC/DMS
characterization of B. megaterium in positive polarity.
FIG. 27 shows the effect of temperature on py-GC/DMS characterization of
B. megater~ium in positive polarity.
FIG. 28 shows the Peak area versus number of bacteria for B. megate~ium
(A), M. luteus (B), and E. coli (C).
FIG. 29 shows overlapping prior art TOF-IMS spectra for m-Xylene and p-
Xylene isomers.
FIG. 30 shows resolved DMS spectra for m-Xylene and p-Xylene.
FIG. 31 shows positive ion spectra for different concentrations of methyl
salycilate.
FIG. 32 shows concentration dependence of the invention to methyl
salycilate for both positive and negative ion spectra.
FIG. 33 shows total ion chromatograms from GC-MS analysis of emissions
of organic compounds trapped on SPME fibers by sampling plumes from
combustion of several materials.
FIG. 34 shows mass spectra from direct characterization of grass, cotton and
cigarette smolce using atmospheric pressure chemical ionization mass
spectrometry.
FIG. 35 shows plots of total intensity of product ions versus retention time
from GC-DMS characterization of emissions of organic compounds trapped on
SPME fibers in plumes from combustion of cotton, paper and grass.
FIG. 36 shows plots of total intensity of product ions versus retention time
from GC-DMS characterization of emissions of organic compounds trapped on
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
_9_
SPME fibers in plumes from combustion of cigarette and engine exhausts. Plots
can
be compared directly to FIG. 35.
FIG. 37 shows topographic plots from GC-DMS characterization of
emissions of organic compounds trapped on SPME fibers in plmnes from
combustion of cotton and paper.
FIG. 38 shows topographic plots from GC-DMS characterization of
emissions of organic compounds trapped on SPME fibers in plumes from
combustion of grass and gasoline.
FIG. 39 shows plots of ion chromatograms extracted from analyses by GC-
DMS of emissions from combustion of cotton (top frame) and paper (bottom
frame).
Ion chromatograms were extracted from plots in FIG. 37.
FIG. 40 shows plots of ion chromatograms extracted from analyses by GC-
DMS of emissions from combustion of grass (top frame) and from engine exhausts
(bottom frame), as extracted from plots in FIG. 38.
FIG. 41 shows resolution of a nerve gas and an interferant simulants at
different radio frequency field strengths by a DMS device of the present
invention.
FIG. 42A and FIG. 42B show DMS spectra of both positive and negative ion
peals, or modes, for a nerve agent stimulant GA.
FIG. 43 shows the effect of reduced pressure on resolution of various peals
by a DMS device of the invention. Resolution increases when pressure is
decreased.
FIG. 44 demonstrates a practice of the invention for a series of warfare agent
simulants selectively mixed with 1% headspace of aqueous fire fighting foam.
As
can be seen, good peak resolution can be achieved in practice of the
invention.
DETAILED DESCRIPTION OF THE INVENTION
The present invention is directed to methods and devices for detection and
identification of analytes in chemical, biological, agricultural or organic
samples
using characteristic mobility-based behavior of the volatilized sample as it
is passed
through a differential ion mobility spectrometry (DMS) device of the
invention. This
characteristic behavior is also referred to herein as a signature, by means of
which
ion species can be separated, detected and identified in practice of the
invention. In
preferred practices, this signature is ion-mobility based and detected in gas-
phase
DMS.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-10-
Preferably each analyte is detected at least by its ion-mobility based
signature. DMS species identification is done by making a species detection
and
comparing this data to stored data which uniquely identify a species based
upon
aspects of ion mobility, i.e., differential mobility. briefly, the data may
include field
conditions (e.g., wavelength, frequency, intensity, among others),
compensation
voltage (a DC offset, or variations in the I~F signal, such as changes in duty
cycle,
among others) and also may include flow characteristics (such as flow rate or
field
gradient, among others) and temperature. DMS produces a signature representing
differences in ion mobility between high field and low field conditions. In
one
embodiment, a signature used in the methods of the present invention is the
combination of the compensation voltage and a field strength that results in a
known
spectral output associated with the species being analyzed. In a further
practice of
the invention, time of flight ion mobility is used to further characterize
aspects of a
detected ion species to further assist species identification.
PRINCIPLES OF DIFFERENTIAL ION MOBILITY SPECTROMETRY
Differential ion mobility spectrometry is a technique for ion separation,
detection and identification. An asymmetric varying high RF field is
established
between filter electrodes over a flow path. Ions in the flow path are driven
by the
field transversely and eventually are neutralized as they contact the
electrodes.
However compensation is applied to return an ion species of interest to the
center of
the flow and to pass through the ion filter unneutralized. This process is
species-
dependent.
If ions derived from two compounds respond differently to the applied high
strength electric field, the ratio of high to low field mobility may be
different for
each compound. Consequently, the magnitude of the compensation necessary to
counteract the drift of the ions toward either plate is also different for
each ion
species. Thus, when a mixture including several species of ions is being
analyzed by
DMS, ideally only one species of ion is selectively transmitted to the
detector for a
given combination of compensation and I~F field. The remaining ions in the
sample
ch-ift toward the filter electrodes and are neutralized upon contact.
The present invention may be practiced with various configurations.
Preferred embodiments feature compact and field-portable, wide-spectrum, dual
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-11-
mode, DMS systems, such as taught in U.S. application number 10/123,030, U.S.
provisional application number 60/389,400, as well as the device described in
U.S.
Pat. Nos. 6,495,823 and 6,512,224. and in copending U.S. Application Serial
IVo.
10/18,4.64., filed 06/28/02, and U.S. Application Serial hTo. 10/4.62,206,
filed
06/13/03. 'The entire teachings ofthe above-referenced disclosures are
incorporated
herein by reference.
APPLIC~TI~1'V ~F TIIE METH~I1S ~F TI3E ~II'~TVEI'~T'rI~1'~1
The methods of the present invention can be used to analyze any volatile or
vohatilizable sample. As used herein, the term "volatile" means evaporating
readily
at normal temperatures and pressures. The term "volatilizable" means capable
of
being converted into gas by use of any of the volatilization methods known in
the
art. As used herein, the term "volatilization" refers to a process of
conversion of
solid or liquid to a gas.
In some embodiments, purification, fractionation and/or separation of a
sample is desired prior to collection of volatile components or volatilization
of a
sample or a part thereof. Samples can be optionally purified or separated
before
beginning the DMS analysis by any of the standard techniques known in the art
such
as HPLC, turbulent flow chromatography, liquid chromatography, reverse phase
chromatography, affinity chromatography, supercritical fluid chromatography,
gas
chromatography (GC), electrophoresis (including but not limited to capillary
electrophoresis, polyacrylamide gel electrophoresis, agarose gel
electrophoresis),
solid phase extraction, and liquid phase extraction, preferably using
different
solvents (e.g., chloroform/methanol for lipids, water for polar molecules).
The
capillary of the ion source could be filled with silica beads (derivatized or
not) or
other material to perform chromatography and/or separation. Volatile or
volatilized
components can further be separated into fractions by any technique known in
the
art, for example, gas chromatography (GC), desorption/absorption, effusion,
condensation, fltration, ion exchange, or the like.
In one embodiment, the sample contains volatile components. A volatile
sample can be collected from the; source by collecting headspace or any other
technique knOWl1 In the ax-t such as f ltration, distillation, sublimation,
vacuum
distillation, etc.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-12-
Volatile or volatilized components can be directly subjected to DMS
analysis or further separated using any of the techniques as described herein.
In one
embodiment, the volatile components are filtered through a membrane to reduce
moisture content and other impuuities that may affect signal-to-noise ratio.
one
spilled in the art can determine the anaterial of a membrane based on the
properties
of the analytes to be separated (for example, polarity). Membrane materials
can
include, for example, polymers such as Teflon or dimethyl silicon.
In other embodiments, samples do not contain volatile components or may
contain a combination of volatile and nonvolatile components. Where analysis
of
non-volatile components of a sample is desired, these samples can be subjected
to
gas-phase DMS analysis as long as they are volatilizable. For example, samples
of
body tissues, pathogens, building materials or samples of soil rnay not be
volatile,
but axe volatilizable. Likewise, breath can contain both volatile components
and
non-volatile but volatilizable components. These components can be separated
as
described herein and the volatilizable components subjected to volatilization
and
analysis by DMS.
The whole sample or any fraction thereof can be subjected to volatilization.
Volatilization can be perfornled in the presence or absence of an oxygen
environment. In one embodiment, volatilization produces a complex mixture of
chemicals referred to herein as "markers". Markers can include whole molecules
or
fragments thereof. The composition and relative abundance of the markers in a
volatilized sample uniquely identifies the sample. Such sample may be organic
or
inorganic, chemical, biological or otherwise.
Any of the techniques known in the art can be used for volatilization.
Preferably, sufficient energy is applied to a sample to break infra- or inter-
molecular
chemical bonds of the analytes in the sample. Non-limiting examples include
pyrolysis, thermal desorption, including temperature-programmed desorption and
thermally assisted solid phase micro-extraction (SPME), laser ionzation,
including
matrix assisted laser desorption ionization (MALDI), microwave excitation
(heating
with microwaves), and chemical transformation (e.g., hydrolysis, photolysis,
oxidation, etc.).
A particularly useful method of volatilization is pyrolysis. The term
pyrolysis (PY) refers to a procedure in which a material is heated, usually in
the
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-13-
absence of oxygen, thereby causing the material to break down into simpler
compounds. Pyr~lysis provides a volatilization technique for various types of
sample analysis, especially for samples that are not easily volatilized.
Pyrolyzing a
sample produces a complex mixture of volatile, semi-volatile and non-volatile
organic chemicals (herein referred to as pyrolysate). vapors generated during
pyrolysis can be swept directly into a detection device. The composition and
relative
abundance of various components in the pyrolysate is a unique characteristic
of a
given sample. Accordingly, the sample can be characterized by analyzing
pyrolysis
products using DIMS to produce a "fingerprint," i.e., signature, that can
uniquely
identify the sample.
Another method for volatilization of a sample is thermal desorption, which is
a widely used technique for extracting and isolating volatile and semi-
volatile
compounds from various matrices. For thermal desorption, samples, usually
solids,
are heated and analytes are volatilized. Typically, a carrier gas or vacuum
system
transports the volatilized components to a detection device. Based on the
operating
temperature of the desorber, thermal desorption processes can be categorized
into
two groups: high temperature thermal desorption (HTTD) (320 to 560°C or
600 to
1000°F) and low temperature thermal desorption (LTTD) (90 to
320°C or 200 to
600°F). It is the technique of choice for air monitoring (indoor,
outdoor, workplace,
automobile interior, breath, etc.) and is a tool for the analysis of soil,
polymers,
packaging materials, foods, flavors, cosmetics, tobacco, building materials,
pharmaceuticals, and consumer products. Almost any sample containing volatile
organic compounds can be analyzed using some variation of this technique.
Temperature programmed desorption (TPD) is a variation of thermal
desorption whereby the temperature of a desorber is increased in a pre-
programmed
manner to maximize the temporal resolution of the analytes and contaminants
(noise) in a sample. TPD is often advantageously coupled to solid phase micro-
extraction (SPIVIE). SPIe~IE is a technique of pre-concentration of analytes
whereby
the analytes of interest are extracted from a sample by absorption into solid
phase
material, usually fibers. Absorption of the analytes by the fibers is based on
the
affinity and solubility of the analytes in the solid phase material of the
fibers. Solid
phase materials can include various polymers, for example, polyacrylate,
polydimethylsiloxane, divinylbenzene and mixtures thereof. Analytes can be
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-14-
extracted from either gas or liquid phases. Alternatively, a solid sample can
be
subjected to thermal desorption and the volatile analytes released during this
process
can be absorbed by SPME fibers, thus pre-concentrating the analytes of
interest.
Following the extraction, the compomids are thermally desorbed by a pre-
y programmed temperature tamping and directed for analysis and detection. An
example of a suitable TPD/SPME device is disclosed in Basile, F.,
~fzstr~ufraentati~ra
Sci. Tech. 31(2): 155-164 (2003), the entire teachings of which are herein
incorporated by reference.
MALDI is a method that allows for vaporization and ionization of non-
volatile samples from a solid-state phase directly into the gas phase at
atmospheric
pressure or in vacuum. Briefly, the technique involves mixing the analyte of
interest
with a large molar excess of a matrix compound, usually a weak organic acid.
This
mixture is placed on a vacuum probe and inserted into a detection device for
laser
desorption analysis. During laser desorption, the matrix that also contains
the
analytes is irradiated with lasers in order to transfer the content into the
gas phase.
The matrix strongly adsorbs the laser light at a wavelength at which the
analyte is
only weakly absorbing. As a result, the matrix reduces intermolecular contacts
beyond analyte-matrix interactions thereby reducing the desorption energy. The
results are high ion yields of the intact analyte and giving rise to sub-
picomole
sensitivity. Principles of MALDI are well-known in the axt. MALDI devices
suitable for use with the present invention are described, for example, in
U.S. Pat.
Nos. 6,414,306 and 6,175,112, the entire teachings of which are herein
incorporated
by reference.
The whole sample or any analyte in a sample can be subjected to
volatilization. In one embodiment, the markers produced by volatilization of a
sample are separated using any of the standard separation techniques used in
the art,
for example, gas chromatography. Following chromatographic separation, any or
all
fractions can be subjected to DMS-based detection.
Various illustrative applications of the methods of the present invention are
described in detail below under separate headings. Generally, samples can be
derived from any source and can include physiological, environmental,
biological,
chemical, agricultural and industrial sources.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-15-
Physiological samples such as breath or tissue samples or physiological
fluids (including blood, urine, synovial fluid, saliva, etc.) can be used to
diagnose
and monitor patient conditions, including point of care patient monitoring,
and
provide forensic infomnation.
Enviromnental samples such as air, soil, sediments, petroleum, natural gas on
water can be used to detect chemical or biological contamination, including
that by
heavy metals, and in monitoring of remediation sites, beans, incinerator
wastes and
water treatment facilities. Methods of the present invention can be used for
detecting agents of chemical and biological warfare in a sample.
Food quality samples, such as foodstuff, air samples from refrigerators or
containers, and swabs of food-contacting surfaces, can be used to detect
bacterial
and chemical contamination of as well as early signs of decomposition during
shipping, in monitoring shelf life and/or pacl~age tampering.
Chemical samples, such as samples of reaction mixtures, can be used to
monitor small and industrial scale processes, including the extent of
reactions and
detection and separation of stereoisomers, as well as to monitor safety
conditions.
Biological samples, such as samples of pathogens, can be used to identify
microorganisms in pure or mixed cultures as well as assess efficiency of
medication
or other antibiotic compounds. Industrial samples, such as samples of
medicaments,
sample of cosmetic products, samples of building materials, samples of crop
plaints,
fabric, synthetic polymers and organic materials, can be used to monitor the
quality
and integrity of the material.
In one embodiment, the sample includes whole microorganisms, non-
microbiotic pathogens or other biological materials. In one embodiment of the
invention, a sample can include protozoan, fungal, bacterial or viral
infectious
agents, antibodies and other proteins, nucleic acids, peptides,
peptidomemetics,
peptide-nucleic acids, oligonucleotides, aptamers, lipids, polysaccharides,
liposaccharides, lipoproteins, glycoproteins, and small molecules. In
preferred
embodiments, the sample contains infectious agents and microorganisms such as
protozoa, fungi, bacteria and virus.
The practice of the method of the present invention includes subjecting
volatile or volatilized marlcers and/or other sample components to DMS
analysis,
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-16-
which can optionally be combined with ion mobility spectrometry (IMS). IMS is
well known in the art.
In alternative embodiments, IMS caaal be used prior to, following or in
parallel with the I~MS analysis. The use of TMS can aid and/or supplement
I~I~1S
analysis in some cases. For example, an I1~S device can be used as an ion
filtor t~
additionally separate and filter the analytes of the sample or the volatilized
markers
and/or analytes, thus raising sensitivity and signal-to-noise ratio of the
I2MS device
and its detectors. Further, a given sample can be subjected to an IMS analysis
in
parallel with a hMS analysis. In this embodiment, IMS spectra can be compared
to
those obtained by a I~MS device. ~y comparing the two types of spectra,
additional
information useful in analysis and identification of the markers and analytes
can be
obtained.
In a further embodiment, the present invention is a method of sample
identification, wherein a sample is analyzed and the analytes contained
therein
identified by a multi-stage process that includes differential mobility
analysis. In
one embodiment, a first stage of sample processing can include filtering a
sample by
particle size, a second stage can include volatilization and the next stage
can include
differential ion mobility. In another embodiment, the present invention is a
method
of analysis of complex mixtures that includes coupled Ion MobilitylMultiple
Stage
Mass Spectrometry.
ILLUSTRATIVE EMBODIMENTS OF THE INVENTION
CLINICAL, INDUSTRIAL, RESEARCH AND PUBLIC HEALTH APPLICATIONS
The present invention can be used in the identification of microorganisms, in
clinical, research, industrial, and public health applications, including
terrorism.
For example, the invention is useful in diagnosing bacterial, viral, fungal
and
protozoan diseases and infections affecting particular patients. Patients can
be
human, primates, companion animals (dogs, cats, birds, fish ete.), livestock
(cows,
sheep, fish, fowl and poultry, etc.). In a preferred embodiment, the present
invention
can be used for pathogen identification in mixed cultures, without the need
for
isolating and culturing of the microorganisms. Briefly, a sample of infected
tissue, a
physiological fluid from a patient or a sample of a pathogen culture is
volatilized.
The pathogen culture can be mixed, i.e., contain more than one type of
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-17-
microorganisms. The volatilized sample, optionally separated by a suitable
separation technique knovrni in the art, such as gas chromatography, is
directed to a
I~1VIS detection device of the present invention. Because the volatilized
sample
comprises markers unique to the pathogens in the sample, the pathogens are
identified. 111 all alternatme embodiment, volatile compounds emitted by the
organisms (methane, ammonia, ethylene, plant alkaloids as well as oxygen,
carbon
dioxide, products of amino acid decarboxylation, protein and lipids
decomposition)
can be detected to assess cell growth and death as well as other physiological
changes.
In one embodiment, the methods of the present invention are used in the
rapid identification of antibiotic resistant strains of organisms such as
Staphylococcus aureus and Mycobacterium tuberculosis, and may optionally be
used to determine to which antibiotics or combinations of antibiotics the pa~.-
ticular
strain of bacterium infecting the patient is susceptible. Additionally, the
invention is
useful in distinguishing between diseases with similar clinical manifestations
but
different causative agents, with possible differences in the preferred course
of
treatment.
In another embodiment, the invention can be used in the identification of
pathogenic and non-pathogenic fungi, as well as determining which agents are
effective against fungal infections. Both the systemic disease caused by
primary
pathogens such as Histoplasma capsulatuzn and the opportunistic mycoses caused
by
Cazzdida albicayzs or Cryptoc~ccus zzeofo~maus can be detected using methods
of the
present invention. Assessment of the type of infectious agent can lead to
better
methods for treatment.
The invention is useful in the diagnosis of at least some of the protozoan
parasites, and especially those, such as N. fowlez~i, which require culturing
for
definitive diagnosis. The invention is also useful in the determination of
appropriate
treatment of an infection by any parasite. For example, in some geographic
areas,
Plaszzzoelizczzz faleipaa~uzz2, the organism associated with the majority of
the one to two
million deaths annually from malaria, has developed resistance to chloroquine,
the
first-line agent used in treatment. As described herein with reference to
bacteria, by
detecting relative abundances of the pathogens in a time series of samples,
the
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-18-
invention can be used to determine to which agents the parasites infecting an
individual show susceptibility.
In another embodiment, the present invention can be used in the diagnosis of
viral diseases, as well as the determination of agents to which particular
viruses
show susceptibility. Exemplary viruses which can be identified using the DMS
analysis of the present invention, either by sampling patient tissues or
bodily fluids
or after replication in culture include: Herpesviruses, which infect
vertebrates,
including humans (Varicella-foster (chickenpox, shingles), Epstein-Barr and
Cytomegalovirus (infectious mononucleosis), Herpes Simplex (herpes, I~aposi's
sarcoma); Baculoviuuses, which infect invertebrates (especially insects such
as silk
worms); iridovirus, which causes African swine fever; Foxviruses, which infect
invertebrates and vertebrates, including humans causing Variola (smallpox),
Vaccinia, Monkeypox, Mousepox; adenovirus, which infect vertebrates, including
humans and causes colds; caulimoviruses, which infect plants; papillomavirus,
which causes warts and other tumors; bacteriophages; hepadnavirus, which
infect
vertebrates, including humans (Hepatitis B); reoviruses, which infect
invertebrates,
plants, and vertebrates, including humans; flaviviruses, which infect
vertebrates,
including humans causing yellow fever; Dengue fever and hepatitis C;
togaviruses,
which infect plants and vertebrates causing rubella, St. Louis,
Eastern/Western
(equine) encephalitis; picornaviruses, which infect vertebrates, including
humans
causing polio, colds and hepatitis A; potyviruses, which infect plants;
various
oncornaviruses, which infect vertebrates, including hmnans, causing cancer
(Avian
Leucosis Virus; Murine Leukemia Virus; Rous Sarcoma Virus; human T-cell
Leukemia Virus (HTLV); Lentiviruses, which infect vertebrates, including
humans,
causing HIV (AIDS) and feline immunodeficiency; orthomyxoviridae, which infect
vertebrates, including humans, causing influenza; filoviridae, which infect
vertebrates, including humans causing Ebola fever, Marburg fever;
paramyxoviruses, which infect vertebrates, including humans (Morbilivirus
(measles), parainfluir~a virus, Rubulavirus (mumps), Respiratory Syncytial
Virus
(colds, croup); rhabdoviridae, which infect invertebrates, plants, or
vertebrates,
including humans (Rabies, Vesicular Stomatitis); arenaviruses, which infect
vertebrates, including humans (Lymphocytic ChorioMeningitis); bunyaviruses,
which infect plants and vertebrates, including humans (La Cross Virus
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-19-
(encephalitis), Sin Nombre Virus (hantavirus pulmonary syndrome), other
hemorrhagic fevers).
The methods of present invention can further be used in determination and
diagnosis of animal and human disease such as Creutzfeld-Jacob disease9
scrapie
and mad cow disease, where the infectious agent is a conformer of a wild-type
analog.
The present invention can be used for public health monitoring. The presence
of coliforms, such as E'. c~li, in waters off beaches is often used as a
marker for the
presence of untreated sewage. The invention can be used in testing of water
samples
to determine whether coliforms are present, without first having to isolate
and
culture the various microorganisms in the sample. Volatilization of a mixed
bacterial sample produces markers unique for each microorganism in a sample.
In
another example, the invention can be used to determine the presence of
Cyyptosporidium or Iribf°io choley~ae in water supplies. The invention
can be used for
testing of municipal water supplies and other waters for the presence of these
and
other pathogens.
The invention can further be used in the determination of which chemical
agents are effective against the particular organism or strain of organism
infecting a
particular patient. Briefly, a time series of samples of infected tissue, a
physiological fluid or samples of a mixed pathogen culture from a patient
being
administered a specific chemical agent are analyzed using aspects of ion
mobility as
described herein. The methods of the present invention can identify which of
the
pathogens responds to the selected treatment. Thus, the invention can be
employed
in screening novel chemical compositions as antibiotics for their potential
efficacy
as agents to bill or inhibit the growth of microorganisms. In general,
screening
typically involves dividing cultures of the organism into multiple aliquots.
The
agents being tested are then introduced into a first group of the aliquots,
while a
second aliquot is reserved as a control to which no agent is added. The
relative
amounts of the organisms in the first test group are then compared to the
amounts of
the organisms grown in the presence of the agents in the second, control group
so
that the relative effectiveness of the agents being tested can be determined.
Large
numbers of aliquots can be cultured and tested in parallel, permitting the
testing of
large numbers of potential agents to be tested at once. Due to high
sensitivity of the
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-20-
methods of the invention, this embodiment can be particularly beneficial where
large
scale culturing of microorganisms is to be avoided.
The general format of microorganism assays employed by the methods of the
present invention will now be described. For ease of descriptican, the methods
described below will be directed toward bacteria, although fungi, protozoa and
viruses can be employed with modifications familiar to persons of skill in the
art.
A specimen to be subjected to analysis, for example blood, urine, mouth or
vaginal swab, personal odor, or a sample of food or of water believed to
contain a
pathogen (~.~: virus, bacteria, fungi or protozoa) is obtained. If not alieady
in the
form of an aqueous suspension, the specimen is usually suspended in an aqueous
medium prior to being subjected to the process of this invention. The size of
the
sample is not critical, provided a sufficient number of microorganisms are
obtained
to permit the intended procedures to be performed. Further, the numbers of
bacteria
present in the aqueous suspension are not critical, provided a sufficient
number of
bacteria are present for the procedure of this invention to detect differences
between
test and control samples. The time required to perform the analysis, however,
can be
reduced as the concentration of microorganisms increases.
If the specimen has too low a cell concentration, it may be concentrated by
known techniques, such as centrifugation or by culturing. The specimen is
cultured
by incubating under conditions suitable for sustaining bacterial growth. The
period
of incubation is that period sufficient to obtain detectable growth, which
will differ
depending upon factors such as bacterial species and concentration of
organisms in
the sample.
After the optional incubation of the test and control specimens for a period
of
time, the specimens are volatilized and quantitatively and qualitatively
analyzed
using ion mobility detection devices of the present invention.
~ne slcilled in the art will readily determine the necessary culturing
conditions. The choice of a particular method of culturing the microorganism
or
microorganisms of interest is determined by a person skilled in the art.
ANALYSIS ~F hHYSIOLOGIC:AL FLUBS AND ~THER ~I~LOGICAL SAMPLES
Analysis of physiological fluids and other biological samples by the methods
of the present invention can be done by employing any of the volatilization
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-21 -
procedures described herein. In one embodiment, the volatile components of a
sample are analyzed by collecting the volatile analytes at a headspace of a
liquid or
solid sample. In another embodiment, a liquid or solid sample can be
volatilized as
described herein.
Samples can be obtained from a variety of sources. ~s will be appreciated
by those in the art, the sources comprise any number of things, including, but
not
limited to, cells (including both primary cells and cultured cell lines),
tissues and
bodily fluids (including, but not limited to, blood, urine, serum, lymph,
bile,
cerebrospinal fluid, interstitial fluid, aqueous or vitreous humor, colostrum,
sputum,
amniotic fluid, saliva, anal and vaginal secretions, perspiration and semen, a
transudate, an exudate (e.g., fluid obtained from an abscess or any other site
of
infection or inflammation) or fluid obtained from a joint (e.g., a normal
joint or a
joint affected by disease such as rheumatoid arthritis, osteoarthritis, gout
or septic
arthritis) of virtually any organism. Samples can be obtained, for example,
from
plants (e.g., crops), invertebrates (e.g., silk worms), non-mammalian
vertebrates
(e.g., poultry, fish, exotic birds, fowl), non-human mammals (e.g., livestock,
companion animals, primates) and humans. Samples can also be collected from
extracellullar fluids, extracellular supernatants from cell cultures,
inclusion bodies in
bacteria, cellular compartments, cellular periplasm, mitochondria compartment,
etc.
In a preferred embodiment, the preparation of samples for the DMS-based
analysis can be achieved by any method known to those of skill in the art.
Sample
preparation can include a desalting step to increase the sensitivity and
resolution. In
addition, as will be appreciated by those in the art, the combination of
preparative
steps, solvents, purification and separation schemes, will all depend on the
classes)
of molecules expected to be detected.
Samples can be prepared in a variety of ways. One skilled in the art will
readily determine the manner of sample preparation, which depends, generally
on
the source and the type of analytes expected to be detected. For example,
physiological fluid samples can be prepared by a protein precipitation
followed by a
desalting treatment. A solution of methanol and water (49:49:2
water:methanol:acetic acid v:v:v) is added to each of the samples and the
samples
are chilled. This precipitates the proteins to the bottom of the tube. Each
tube is then
centrifuged and the supernatant decanted. For the desalting step, small amount
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
_22_
(approximately 100 mg) of DOWER ion exchange resin is added to each vial and
allowed to sit for approximately 10 minutes. The sample is then centrifuged
and the
supernatant removed. This soluti~n is then introduced to the an ion-mobility
detection device.
Any solvent l~aown to those of shill in the art can be used in conjunction
with an ion source in the practice of the present invention. Examples of
suitable
solvents are dimethylsulfoxide, acetonitrile, N,1V-dimethyl formamide,
propylene
Garbozlate, methylene chloride, nitromethane, nitrober~ene, hexane, methanol
and
water. The solvent can comprise more than one solvent. In a preferred
embodiment,
the solvent is a solution of methanol and water (49:49:2,
water:methanol:acetic acid
v:v:v). Selection of a suitable solvent will depend on the type of molecules
is
expected to be detected. For example, a solution of methanol and water is used
as a
solvent when the detection of soluble molecules is to be achieved by the ion-
mobility device, while hexane can be used when the detection of apolar
molecules
such as lipids is to be achieved. In one embodiment of the invention, the
sample
source (e.g., tissues, cells) is extracted in different solvents and each
extraction
subjected to analysis, so that a more complete analysis ofthe molecules
present in
the sample source can be accomplished.
Samples can optionally be purified, fractionated and/or separated using any
of the standard technique known in the art as described herein.
W addition, it should be noted that purification and separation techniques
may be simultaneously or sequentially run on samples, in different orders and
in
different combinations. Thus for example, a simple protein precipitation may
be nm
on a portion of the sample, and then a HPLC step. Similarly, portions of
samples
(e.g., portions of the cellular populations) may be subjected to different
techniques
in the elucidation or identification of peaks.
While one spilled in the ant will appreciate that the method of the present
invention can be practiced on a sample obtained from any of the sources
mentioned
above, application of ion-mobility analysis to detection and identification of
urine
constituents will now be described in details.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
- 23 -
URINALYSIS
An embodiment of the present invention provides a method of urinalysis
using aspects of I~I~1S lOn 111~blhty ~f the ~,1r111e ~~n~tltLle11t8.
Urinalysis is an examination of the urine by physical or chemical means.
Urinalysis comprises a battery of chemical and microscopic tests that help to
screen
for urinary tract infections, renal disease, and diseases of other organs that
result in
abnormal metabolites appearing in the urine.
The following is a non-limiting list of indicators present in urine can be
detected: bilirubin (a degradation product of hemoglobin); glucose; hemoglobin
(an
indication of hemolysis); urine ketones (a by-product of fat metabolism and
present
in starvation and diabetes); nitrite (an indication of urinary tract
infection); urine pH
(the acidity or alkalinity of the urine); urine protein; urobilinogen (a
degradation
product of bilirubin). In addition, pathogens such as fungi (yeasts),
protozoa,
bacteria and virus can be detected.
The results of the analyses are used for diagnosing the patients.
Specifically,.
in some situations, an alkaline urine is good. Kidney stones are less likely
to foam
and some antibiotics are more effective in the alkaline urine. There may be
times
when the acidic urine may help prevent some kinds of l~idney stones and may
prevent growth of certain types of bacteria. When blood levels of glucose are
very
high, some of the glucose may show up in the urine. The glucose and the
ketones
tests are usually done together. Large amounts of ketones may be present in
uncontrolled diabetes. Finding protein in the urine is probably the best test
for
screening fox kidney disease, although there may be a number of causes for an
increased protein level in the urine. Bilirubin in the urine is a sign of a
liver or bile
duct disease. Urobilinogen is found in small traces in the urine. Nitrites and
white
blood cells are an indication that a urinary tract infection is present. Any
vitamin C
that the body does not need is excreted in the urine. If there are measurable
amounts
of Vitamin C in the urine, it may interfere with the other urine tests.
Volatile components of a urine sample are colected by directing a headspace
to a DMS detection device. In another embodiment, urine sample is subjected to
gas-chromatographic separation prior to DMS analysis. In an exemplary
procedure,
a gas chromatograph is maintained at an initial temperature designated as To.
At the
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-24-
onset of analysis, designated as time to, the sample is introduced to the
inlet of the
gas chromatograph column. The temperature of the gas chromatograph is then
elevated or ramped at a constant rate to a temperature Tr, reached at time tr
at which
all analytes have completed elution from the c~lunm. The column is further
heated
to a final temperature Tf, slightly elevated above Tr, at time tl, and is held
at this
temperature to clean out the column. At the end of this final period,
designated as
time t2, the chromatograph is cooled back down to the initial temperature To
for
subsequent analysis, which cooling down is completed at a time t3.
In addition to the above, every person emits a particular chemical odor,
which may be used as a signature. This odor signature can be detected by DMS
practices of the invention. The odor signature can be used to identify an
individual
for security applications, to identify bodies, and for covert applications to
determine
if a terrorist has been residing in a particular location based on signatures
from twine
or other residual odors left at the location. Such signature detection, based
on
detection of volatile chemicals, may be practiced according to the present
invention.
BREATH ANALYSIS
An embodiment of the present invention provides a method of breath
analysis. The present invention provides a method and apparatus that can
measure
and analyze both volatile and relatively non-volatile components released.
Alcohols,
such as ethanol, can be detected in exhaled air. Furthermore, understanding
the
composition of breath analysis can be used to diagnose diseases and elucidate
pharmacokinetic properties of various compounds.
Alveolar breath is a distinctive gas whose chemical composition differs
markedly from inspired air. Volatile organic compounds (VOCs) are either
subtracted from inspired air (by degradation and/or excretion in the body) or
added
to alveolar breath as products of metabolism. I~Tornal human breath contained
several hundred different V~Cs in low concentrations. More than a thousand
different V~Cs have been observed in low concentrations in normal human
breath.
(Phillips M: Method for the collection and assay of volatile organic compounds
in
breath, Ari.alytieal Bi~elzerraist~y 1997; 247:272-278, the relevant parts of
which are
incorporated herein by reference).
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
- 25 -
Alkanes in breath are markers of oxygen free radical (OFR) activity in vivo.
OFR's degrade biological membranes by lipid peroxidation, converting
polyunsaturated fatty acids to alkanes which are excreted through the lungs as
volatile organic compounds (S,OCs); (I~eephen s C. M. F., et al., The hydr
ocarbon
~ breath test in the study of lipid peroxidation; principles and practice,
Clina. ~yavest.
ll~led. 1992; 15(2):163-186).
For example, increased pentane in the breath has been reported as a marker
of oxidative stress in several diseases including breast cancer (Hietanen E.,
et al.,
Diet and oxidative stress in breast, colon and prostate cancer patients: a
case control
study, Eu~opeafz JouYnal of Clinical Nutrition 1994; 48:575-586), heart
transplant
rejection (Sobotka P. A., et al., Breath pentane is a marker of acute cardiac
allograft
rejection. J. Heat Lung Transplant 1994; 13:224-9), acute myocardial
infarction
(Weitz Z. W., et al., High breath pentane concentrations during acute
myocardial
infarction. Lancet 1991; 337:933-35), schizophrenia (Kovaleva E. S, et al.,
Lipid
peroxidation processes in patients with schizophrenia. Zh IVevropatol Psikiatr
1989:
89(5): 108-10), rheumatoid arthritis (Humad S., et al., Breath pentane
excretion as a
marker of disease activity in rheumatoid arthritis, Free Rad Res ConanZS 198;
5(2):101-106) and bronchial asthma (Olopade C. O., et al., Exhaled pentane
levels in
acute asthma, ClZest 1997; 111(4):862-5). Analysis ofbreath alkanes could
potentially provide a new and non-invasive method for early detection of some
of
these disorders (Phillips M: Breath tests in medicine, Scientifzc American
1992;
267(1):74-79).
Breast cancer can be detected by identifying metabolic products of the
cytochrome P450-mediated pathways. The cytochrome P450 (CYP) system
comprises a group of mixed function oxidase enzymes which metabolize drugs and
other xenobiotics. This system also metabolizes alkanes to alcohols e.g. n-
hexane to
2- and 3-hexanol. The cytochrome P450 system is reportedly expressed in
cancers
of breast as well as other tissues (Murray Ca. L, et al., Tumor-specific
expression of
cytoclarome P4~50 C~YP1B1. Caaacei~Res 1997; 57(14.):3026-31). Recent studies
suggest that exhaled pentane can be used as an additional marker for breast
cancer.
Hietanen et al. studied 20 women with histologically proven breast cancer and
a
group of age and sex-matched controls (Hietanen E., et al., Diet and oxidative
stress
in breast, colon and prostate cancer patients: a case control study, European
.Iou~nal
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-26-
of Clinical Nut~itiofz 1994; 48:575-586). Mean breath pentane concentration in
the
cancer patients (2.6 ppb, SIB=2.8) was significantly higher than in the
controls (0.6
ppb, SD=1.1, p<0.01). They did not report concentrations of pentane in ambient
air,
nor the alveolar gradients ofpentane.
The methods and devices of the present invention can be particularly useful
in diagnosing ischemic heart disease. There is an increasing body of evidence
that
myocardial oxygen free radical activity is increased in ischemic heart
disease.
~xidative stress also increases during surgical reperfusion of the heaut, or
after
thrombolysis, and it is related to transient left ventricular dysfunction, or
stunning
(Ferrari R.; ct al., ~xidative stress during myocardial ischemia and heart
failure, Eur
Heart J 1998; 19 Suppl B:B2-11). Pentane was significantly increased in 10
patients
with acute myocardial infarction compared to 10 healthy controls (Weitz Z W,
et al.,
High breath pentane concentrations during acute myocardial infarction. Lancet
1991;337:933-35). However, a fundamental flaw in the conventional breath
pentane
assays is that the colunnl employed in the gas chromatograph does not separate
pentane from isoprene, the most abundant compound in breath. The devices
employed by the methods of the present invention can separate pentane and
isoprene
from one another
Methods of the present invention can be used as non-invasive techniques to
diagnose organ rejection, including heart. There is a well-documented
biochemical
basis for breath testing that provides for the early detection of transplant
rejection.
Tissue damage arising from inflammation is accompanied by an accumulation of
intracellular oxygen free radicals (OFR'S) which cause lipid peroxidation of
lipid
membranes (I~neepkens C. M. F., et al., The hydrocarbon breath test in the
study of
lipid peroxidation: principles and practice. Clirt Invest Med 1992; 15(2):163-
186.
Kneeplcens C. M. F., et al., The potential of the hydrocarbon breath test as a
measure
of lipid peroxidation. Fy~ee Radic Biol ll~Ied 1994; 17:127-60). This process
is
accompanied by the evolution of alkanes which are excreted in the breath. ~ne
of
these allcanes, pentane, is the best documented marker of ~FR activity.
Methods of
the present invention can be employed to detect breath pentane in tratlsplant
recipients.
End-stage renal disease (ESRD) is a fatal condition unless it is treated with
either lcidney transplantation or dialysis of blood or peritoneal fluid.
Clinicians who
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
_27_
come into contact with patients with chronic renal failure are familiar with
the
classic odor of uremic breath. It has been variously described as "fishy",
"armnoniacal" and "fetid". This odor arises from presence of trimethylamine
and
dimethylamine in the blood, as well as in creased con centrations of secondary
and
tertiary amines. Methods of the present invention can be used to detect these
compounds, thus indicating the presence of ESI~D.
Additionally, presence of methylated alkanes are common components of the
breath in normal humans as well as in those suffering from lung cancer.
Phillips M.,
et czl., Variation in volatile organic compounds in the breath of normal
humans.
Journal o~''Ch~orncztog~czphy ~ 629 (1-2): 75-~~; 1999; Phillips M., et czl.,
Volatile
orgasuc compounds in breath as markers of lung cancer: a cross-sectional
study.
Lancet 353:1930-33; 1999. These VOCs appeared to provide additional markers of
oxidative stress. Methods and devices employed by the present invention can be
used to create patient's methylated alkane profile and thus serve as
additional tool in
early lung cancer diagnosing.
Various personal odors may be detected in practice of the invention, such as
breath or arm pit or the like. In another embodiment, the present invention is
a
breath test which can be used to determine the characteristic of metabolism of
a drug
in a subject. Specifically, the present invention can be used to determine
this
characteristic of metabolism by measuring the concentration of a metabolite in
the
exhaled breath of the subject after an appropriate amount of the drug has been
administered to the subj ect. Hereinafter, the term "characteristic of
metabolism"
includes whether such metabolism occurs, the rate of metabolism and the extent
of
metabolism. However, for clarity the aforementioned and following descriptions
specifically describe the measurement of the rate of metabolism. Generally,
these
tests involve the administration of a substrate to the subject and the
measurement of
one or more cleavage products produced when the substrate is chemically
cleaved.
For example, detecting t~Ielicobczcter pylori, which produces a large quantity
of
the enzyme urease, can be accomplished by orally adminstering urea to a
subject
with subsequent monitoring of the exhaled dioaxide and ammonia.
breath tests can be used to measure physiological processes such as the rate
of
gastric emptying. For example, gastric emptying rates were measured for solids
and
liquids by using octanoate or acetate as the substrate (Duan, L.-P., et al.,
Digestive
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-28-
Diseases ahd Sciehces, 1995, 40:2200-2206). The substrate can be administered
to
the subject and the exhaled breath of the subject was measured with an ion-
mobility
detection device.
The breath test of the present inventi~n can be performed as follows. First,
the drug is administered to the subj ect. Next, the exhaled breath of the subj
ect is
analyzed after a suitable time period for a concentration of a metabolite of
the drug,
the concentration indicating the rate of metabolism of the drug in the subj
ect. Such a
breath test has a number of advantages over conventional methods for
determining
the concentration of a drug in the subj ect. Not only is a breath test non-
invasive, it is
also more rapid than analyzing blood samples and it can also be performed
multiple
times on the subject.
In one embodiment, the present invention provides a method and apparatus
that can measure and analyze components released from food or other products
during oral processing. An optional step of volatilization is employed should
detection of non-volatile components is desired. This method can be used to
measure
components that are present in small concentrations, yet are important to the
flavor
of a product.
Illustrative embodiments of breath sample collection techniques will now be
described. In one embodiment, the collection system included a mouthpiece and
a
tube for carrying the exhaled breath of the subject into a mixing chamber. A
sample
outlet and exit tube connected the mixing chamber to a measuring device, such
as
DMS detection device. W addition, heat can be applied to the system to prevent
condensation of moisture from the breath sample on the system components. The
mixing chamber can be provided to insure that the exhaled breath sample mixed
with previous samples, and that a small quantity of the combined breath sample
was
drawn from the chamber into the measuring device.
In another embodiment, the collection system uses a pumping system to
draw an air sample from the nose of a subject, through a nose-piece and past a
membrane separator fitted to a DMS detection device.
In yet another embodiment, a carrier gas, e.~. N2, and a breath sample can be
injected into a separating column, e.g~. Ca~, and then circulated toward an
ion-
mobility detection device.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-29-
In smother embodiment, the collection system described in IJ.S. Pat. No.
5,479,815. In this system, the subject exhales a breath sample into one or
more
collection chambers that are preferably maintained in a temperature controlled
cabinet to prevent condensation of pouions of the sample. Either month-
e~~haled or
nose-exhaled air can be collected. Where food flavorings are being detected,
samples are preferably collected fiom the subject's mouth. Each breath sample
can
be purged from its collection chamber with a non-reactive gas flow into a trap
containing an interface that separates and collects components from the breath
sample. This interface preferably is an inert adsorbent material selected for
its ability
to. A substrate coated with an absorbent or other material capable of
collecting
components may also be used. The interface, in addition, preferably permits
any
moisture contained in the breath sample to pass through the trap, leaving only
the
collected components on the interface.
The adsorbent trap is then transferred to a thermal desorber or other device
capable of releasing the adsorbed components from the interface surface into a
measuring and analysis apparatus, such as an ion-mobility based detection
device of
the invention. In one embodiment, the components collected by an adsorptive
trap
are thermally desorbed from the trap and flushed into a gas chromatograph by a
non-
reactive gas flow prior to being identified by an ion-mobility detection
device.
In another embodiment, the subject blows each breath sample first through a
condensation trap and then into the collection chamber. The condensation trap
captures relatively non-volatile flavor components that might not otherwise be
recovered from the collection chamber or could not,be readily desorbed from an
adsorbent trap. The condensation trap preferably includes non-reactive glass
tubing
pacl~ed with a non-reactive and non-adsorbent substrate, such as glass wool.
Other
materials may be used provided that the materials withstand high temperatures
and
do not react with flavor compounds. The trap is maintained at a temperature
that will
encourage the condensation on the substrate of slightly volatile or relatively
non-
volatile flavor components in the subject's breath and permits the volatile
components to pass through into the collection chamber.
The condensation trap is then heated to re-vaporise the condensed components,
so that they may be purged into measuring and analysis devices such as a gas
chromatograph and mass measuring device. Similarly, the flavor components that
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-30-
passed through the condensation trap into the collection chamber also may be
flushed into an adsorbent trap so that they may be studied as discussed above.
~tJALIT1' C~NTR~L QF FCaODSTLTFF AID F~~I~-hI2~CESSINE"a SUI~.F.~~ES
The present invention can be used in the detection of chemical contaminants
and/or viable pathogens that may be present in processed foods, such as found
in
ground beef and other meats. The ability to rapidly confirm or disprove the
presence
of significant contamination may, for e~~ample, reduce or eliminate the need
for
destruction or recall of ground meats or other foods in cases where
contamination
was possible but not certain by demonstrating that the meats or other foods in
question are not contaminated with viable bacteria at the time of testing.
The present invention can be used in food processing plants, hospitals,
laboratories, and other facilities to determine whether surfaces are free of
pathogens
or whether additional or more stringent sterilization or containment
procedures are
required. This can be achieved by, for example, collecting a sample of dust or
a
cotton swab of a surface in question, followed by either re-suspending the
sample
material in a fluid or immediate volatilization. The volatilized sample is
subjected to
detection as described herein.
The present invention can further be used to identify contaminated food
containers by collecting either a sample of a headspace or a swab of a
surface.
In one embodiment, the present invention can be used for detection of
substances in a foodstuff industry. Both odorous (relatively volatile) and non-
odorous substances (relatively non-volatile) can be detected, the latter with
an
additional optional step of volatilization being required before the sample
being
directed into a DMS device.
The detection of odorous substances has many industrial applications,
especially in processes in the foodstuff industry, in which one can, for
example,
determine the degree of freshness and the quality of the products, due to the
odorous
substances which they release. Cas chromatography, which consists in a method
of
separating the molecules of gas compositions, can be used as a way of pre-
separating components of a sample prior to detection by DMS.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-31-
A method for detecting odorous or volatile substances or substances made
volatile, comprises the steps of directing an air sample collected in the
proximity of
a food substance or a food-handling surface to DMS detection device.
This method can advantageously employ sa.~nple collection teclataiques
described in LT.S. Fat. No. 5,~01,2~7. In particular, the transport of the
odorous
substances from a sample can be achieved by a variable controlled flow of gas.
This
allows for very rapid detection of odorous substances, in about a few seconds.
The
means used in order to achieve this variable controlled flow of gas are
advantageously made up of at least a pump with variable flow rate. During the
phase
of separating and detecting the components of a sample, the gas flow rate can
be
diminished or increased according to whether the substances become more or
less
volatile throughout the duration of the measurement period.
SMART DMS SMOKE DETECTOR
The present invention can be used for identification of a fuel source by
analyzing smoke. Where a gas chromatography device is used to pre-separate the
smolce components, GC-DMS instrument operating at ambient pressure in air
provides a compact and convenient fuel-specific smoke alarm at a reasonable
cost.
The specificity and sensitivity of the system earns the moniker of a "smart
DMS
smoke detector".
One embodiment of the present invention provides a smart smoke detector
with' high specificity by detection of volatile organic constituents (VOCs) in
the
smolce. Fires produce a large number of organic compounds in complex mixtures
in
the vapor phase as seen in fixes from synthetic polymers, cigarettes, and
cellulose-
based materials such as wood or cotton. This approach provides sufficient
analytical
information for selective detection of fire components through measurement of
the
chemical composition of the emissions.
1LLLUSTItATIVE ~IFFEI~I'~ITIAI,1L~I'~T 1~V~IJ ~EILITY DEVICES ~F TIIE
l~IZESE~~1T
~NVE~TI~1~
An illustrative practice of the invention is shown as system 10 in FIG. 1, in
which a sample A and transport fluid B axe delivered to a filter C (operating
by
aspects of ion mobility), wherein the ionized sample S+~- is filtered by ion
species
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-32-
according to aspects of ion mobility. Thus a selected ionized species is
outputted
from the ion filter ("separator") and may be further processed in part D. This
further
processing may include being detected to indicate presence of a biological
material
in the sample, and/or may include being collected and uved as a biological
material
extracted from the sample. The sample itself may be delivered already
including ion
species of interest.
The sample is iouzed in ionization region E before it enters the filter part
C.
In one alternative embodiment, region E receives the sample and transport
fluid
where they are mixed together in mixing region E1 in presence of an ionization
source E2, or are mixed with an additional ionized fluid flow from source E2,
all to
provide the ionized sample S+~- that is delivered from region E to f lter C.
Filter C is
preferably a DMS filter.
In a preferred illustration of the invention, an analyte is detected based on
differences uz mobility of the ionized analyte in a DMS electric filter field.
Preferably, this field analysis includes high field asymmetric waveform ion
mobility
spectrometry-type differential ion mobility, as described in U.S. Pat. No.
6,495,823
or U.S. Pat. No. 6,512,224, and generally described herein as DMS. These
patents
teach both gas transport of ion species and electric field ion transport of
ion species,
which may be practiced in embodiments of the present invention.
In one practice of the invention, a DMS filter (separator) is tuned to pass a
specific analyte of interest, and the passed analyte is then collected or
processed
accordingly. Tn another practice, the filter field is scanned through a range
that
enables detection and identification of a range of ion species that are
present in the
sample, including positive and negative species. This spectrum can be detected
for a
full characterization of the detected sample.
A fiu-ther practice of the invention is shown in FIG. 2A wherein system 10
includes separation sections "S-A" and "S-B" followed by identification
section
GGm99. ~ operation a complex. sample S~' can be separated in first section S-A
and
thlJe~separated sample flow S is then applied to the second section S-B for
further
separatiouprocessing. The result of both separations and the related S-A and S-
B
data is correlated and enables reliable identification of separated components
from
the complex sample. The output of section S-B is evaluated in the
identification
section ID.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
- 33 -
It will be understood that a preferred process of the invention includes using
DMS to generate separation data and at least one other processing step that
yields its
own separation data. This additional separation step may be bef~re or after
DMS
filtering. The combination ~f detection data leads to highly reliable
identification of
ion species present in a complex sample even at trace levels. In a preferred
embodiment, the first separation section s-A includes a gas chr~matograph (GC)
and the second section is a DMS filter.
Accordingly, in one embodiment, the present invention is an apparatus and a
method for detection and identification of analytes in a sample by aspects of
ion-
mobility based detection. In this embodiment, a portion of a sample is
directed into
a first separation device, thereby obtaining a first profile of a sample. A
portion of a
sample is also directed into a second separation device thereby obtaining a
second
profile of a sample. At least one of the first and the second separation
devices is a
DMS device. As used herein, a "profile" includes any data obtained by a
separation
device, such as, for example, any ion-mobility signature, such as DMS
mobility,
time of flight, mass spectra, chromatographic retention time and the like. One
skilled in the art will determine specific data comprising a profile based on
the
nature of the sample to be analyzed and the separation device employed by a
specific embodiment of the apparatus and a method of the present invention.
The first and the second profiles obtained above are combined, thereby
allowing identification of at least one analyte in a sample. The combination
of
profiles can be done by way of comparison of the two profiles, whereby the
presence
of a particular analyte can be confirmed. The combination can further include
adding the data obtained by a first separation device to the data obtained by
the
second separation device.
The DMS practice of the invention may include a DMS filter in either or
both separators S-A or S-B. A preferred DMS system is shown in FIG. 2B,
including a sample input section 10A, a I~Ii~TS ion filter section 10~. The
DMS
output is evaluated in the detection and identification section l OC in FIG.
2~ (which
constitutes the identification section ~ of Figure 2A), wluch includes an
intelligent
controller/driver provided by command and control unit 34.
Typically, a memory or data store 33 is used to record separation or mobility
signatures for known ion species and the apparatus is enabled with this data.
The
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-34-
DMS detection data can be correlated with the field conditions data (e.g., RF
characteristics, compensation, flow rate, temp., etc.) and forms a detection
dataset
for the detected ion. The detection dataset is compared to a signature
dataset(s) for
l~nown ion species as stored in the data store. A match enables identification
of the
species of the detected ion. In one emb~diment~ the data store includes
associated
characteristic retention time data and data that relates to use of other
separation
techniques (such as a thermally controlled SPME prefilter).
The detection is both qualitative axed quantitative. For example, upon
detection of an ion associated with an anthrax molecule, a match with the
stored data
will enable identification of the detected species as "anthrax" and with an
indication
of detection level based on the intensity of detection. Such an indication may
be
issued as a warning to a display or other output device.
The DMS RF signal is generated and the compensation bias is applied to
filter electrodes 20, 22 by drive circuits 32 within command and control unit
34.
Preferably a detector 26 is provided, and preferably a charge detector (e.g.,
Faraday
type detector), including detector electrodes 28, 30. As ions contact the
detector
electrodes they deposit their charges and these detection signals are then
amplified
by amplifiers 36, 38, all under direction and control of unit 34. Preferably a
computer or microprocessor 40 correlates drive signals applied to the filter
electrodes with detection signals from amplifiers 36, 38, and makes a
comparison to
stored data in data store 41, and then issues identification data 42 to a
readout
device, such as for indication of detection of the target molecule.
In one embodiment, a GC output of section S-A delivers sample S as an
eluant that is carried by a drift or carrier gas G into flow channel 11 at
inlet 12. The
sample S flows toward a sample outlet 13 at the other end of flow channel 11.
Sample S may include various molecules including trace level analyte T. The
sample may be delivered directly from the GC (or in other embodiments may be
delivered via a nebuiizer, spray head, ete.), and flows into ionization region
I4.
Molecules irl the sample axe then ionized by ionization source 16.
The result of ionization is ionized analyte ions T+/- and other ions S+, S-
with some neutral molecules S°, as may be derived from various chemical
species
that are in sample S. These ions may appear as monomers, dimers, clusters,
etc.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-35-
The ions and neutral molecules is flowed into the ion filter section l OB for
analysis. The Garner gas carries the ionzed sample into the analytical gap 19
formed
between electrodes 20, 22 of filter 24. In a preferred embodiment of the
invention,
filtering proceeds based upon differences in ion zn~bilit~r in aai asymmetric
IZf filter
field alternating between high and low field values. This filtering reflects
unique
mobility characteristics of the ions as species; the process enables
discrimination of
species based upon mobility characteristics in tlae field which reflects ion
size,
shape, mass, charge, etc.
In accordance with an illustrative embodiment of the present invention, an
asymmetric field voltage, Vrf, applied across the filter electrodes 20, 22
generates a
field F (e.g.,10,000 V/cm) whose strength alternates between high value Vmax
and
low value Vmin. This variation in the field strength causes the ions to move
transverse to the sample flow in the flow channel, with the transverse motion
being
representative of the characteristic field mobility of the ions.
The mobility in the high field condition differs from that of the low field
condition, and this mobility difference produces a net transverse displacement
of the
ions as they travel longitudinally through the filter field between the
electrodes,
resulting in an ion traj ectory over time. This traj ectory will drive the
ions into the
filter electrodes, causing them to be neutralized, lacking a countervailing
compensation.
A compensation, such as a DC compensation voltage Vcomp, is applied to
the filter to differentially compensate this transverse motion. The
compensation will
compensate the transverse motion of a selected ion species and will cause it
to return
to the center of the flow path based on its compensated mobility
characteristics.
Thus this returned ion species is able to exit the filter without colliding
with the filter
electrodes and without being neutralized.
In this process other species will not be sufficiently compensated and will
collide with the filter electrodes 20, 22 and will be neutralized. The
neutralized ions
T~ are purged by the carrier gas, or by heating the flow path 11, for example.
A compomad may be represented by either or both positive and negative ions
("modes") such as T+ and T-, as such modes may be generated by ionization of
the
analyte molecules T. hl a preferred embodiment, both positive and negative
modes
of an ionized species can be simultaneously detected in detection and
identification
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-36-
section l OC. In this case, detector 26 includes biased detector electrodes
28, 30 that
are capable of simultaneous detection of modes simultaneously passed by the
DMS
filter.
The in-line configuration of the fl~v~r path enables both modes caf a sp~;cies
generated during ionisation to flow into the 1~MS filter 24.. The I~1~I1S
filter passes
these modes during a mobility scan, where each is the passed species when the
scanned field conditions are appropriate. Thus analyte T may produce ions T+
and
T- which each will pass through the filter at the appropriate signature field
conditions.
In practice of the invention, ion species are filtered based on mobility
differences. Therefore in a preferred practice of the invention, ail ions of
an ion
species will be passed on for detection, whether positive or negative ions,
and which
may be detected simultaneously. Accordingly the detector electrodes are biased
so
that one attracts the positive and the other attracts the negative ions. Thus,
in an
embodiment of such arrangement both "positive mode" and "negative mode" ions
of
a species are detected simultaneocxsly. Having both modes from a sirtgZe
detection
provides a more unique signature for the detected ion species and therefore
increases
the potential accuracy of species identification of the invention. The benefit
of mode
detections is further discussed below.
In a practice of an embodiment of the invention, the ions flow to the detector
wherein electrode 28 may be biased positive and electrode 30 biased negative,
and
therefore electrode 28 steers the positive ions T+ toward electrode 30, and
results in
ions T+ depositing their charges on electrode 30. Meanwhile, electrode 30 acts
as a
steering electrode and steers the negative ions T- toward electrode 28, and
results in
ions T- depositing their charges on electrode 28. It is a feature of this
embodiment
that both + and - ion modes may be detected simultaneously. Single mode ar
dual
mode detection data is combined with filter field parameter data and is then
compared to stored data to male an identification of the detected analyte T,
and this
is combined with the separation data. representing the first separation to
enable
highly reliable identification of the analyte of interest, even at trace
levels.
In accordance with the present invention, discrimination of ions from each
other according to mobility differences is achieved wherein the RF field and
the
selected compensation enables a particular ion species to pass though the
filter. A
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-37-
plot of detection intensity versus compensation for a given field strength is
shown in
FIG. 2C, where peaks Ta, Tb indicate intensity of the detection signal at
compensation levels a, b for the particular RF field. Peak "Ta+" represents
the
"analyte Ta" positive mode, and peak "Tb+" represents "analyte Tb" positive
mode.
Peak "Tb-" represents "analyte Tb" negative mode. The intensity of the peak
may be
correlated with detection quantity. Furthermore, the retention time associated
with
these pealcs can be correlated with the peaks to improve reliability of
species
identification.
In a simplified aspect of the invention, a first separator S-A (e.g., a fast
GC)
is coupled with to a second separator S-B (e.~., I)MS filter), as in the
arrangement of
FIG. 2A and 2B. In a preferred embodiment, a pre-filtering step is provided by
a
front-end collector system S-C (shown in dotted outline) to enable a lughly
reliable,
selective and sensitive advanced chemical detector system. In practice of the
invention, various analytes that are difficult to discriminate and detect can
be
identified with confidence.
For example, the sample can be collected, such as by means of solid-phase
micro-extraction (SPME) media, in pre-collector part S-C, and then delivered
to a
GC, in separator S-A, for further separation and followed by DMS separation in
part
S-B and species identification in part TD, according to the invention.
SPME uses a fiber or tube having coating material which preferentially
adsorbs analytes from a sample matrix and delivers the analyte for further
processing. SPME is routinely applied to gas-phase liquid-phase extractions,
such
as fox extracting organic analytes from a sample and delivering same for
chromatographic analysis. The preferred embodiment enables delivery of a
volatilized or volatilizable sample which can be processed in gas phase.
Embodiments of the present invention enable provision of intelligent
monitors for a variety of application. For example, smart air monitors, smoke
detectors and the lilce can be provides. Prograammable control of the
separation
secti~ns or selection of dedicated components enables tailoring a system to
paa.-ticular needs. An illustrative embodiment of this invention enables an
advanced
environmental detector of reasonably high analytical performance. In this
embodiment, a SPME collector system 100, shown in FIG. 3, which provides a pre-
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-3S-
filter front-end (see separation section S-C) and delivers sample S to the GC
(separation section S-B) and then the DMS section S-B proceeds as described.
The SPME collector system i00 preferably has several SPME collectors 101,
102, 103 ... n. which are selected for special characteristics. For example,
all may
be designed to selectively deliver a particular type of analyte or a range of
analytes
into collector 110.
The SPME fibers introduce the sample S into collector 110. In one
embodiment, a drift gas 112 is introduced into collector 110 and carries the
sample S
introduced by the SPME fibers to the separator S-A (~.g., GC). 1lTow further
separation, filtering, detection and identification proceeds as earlier
described.
This SPME sampling may include the step of heating the SPME fibers to
purge VOCs, or to pyrolyze/volatilize a sample that then is delivered into
collector
114 anal then for further processing such as by GC-DMS. The heating may be
provided by heaters 116 associated with the SPME absorber fibers 101, 103,
105,
etc. The system is controlled by controller 34.
The heaters may be switched on and off and in this manner carp control
sample delivery within a desired range of chemical compounds according to the
characteristics of the switched SPME fibers. For example, upon detection
signal, one
or a series of the fibers 101-n can be heated to change the sample absorption
profile
and delivery characteristics. Heaters 116 can also be used to heat the fibers
for
purging of same or even for pyrolysis of the samples. Such heating can be
ramped to
create a desired profile.
In one embodiment, the present invention provides detection of smoke and
intelligent discrimination of components in the smoke. The chemical
composition of
smoke from sources of interest exhibit measurable chemical differences that
can be
analyzed by the SPME-GC-DMS system. A remarkable level of reproducibility for
complex chemical process (i.e., combustion of natural materials) was obtained
using
simple sampling methods. In a further embodiment, monitoring changing vapor
composition with time of bum and detailed identification of volatile products
from
combustion is used in an augmented method of the invention.
Tt will be appreciated that the present invention is clot limited to
detection/discrimination of smoke. More broadly, the present invention enables
analysis of compounds by including differential ion mobility analysis in
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-39-
compensated high field asymmetric waveform ion mobility RF fields in a compact
paclcage that can be manufactured using high volume techniques that result in
low
per costs and yet produces results comparable to expensive analytical
equipment.
Systems accordiiag to the invention can be light-weight and yet provide the
ability to
provide highly effective analytical equipment whether in the field or in the
laboratory.
Devices of the invention are able to rapidly produce accurate, real-time or
near real-time, in-situ, orthogonal data for identification of a wide range of
chemical
compounds. Devices of the invention, such as devices according to FIG. 2A
through
2C and FIG. 3, enable distributed installation of detection systems such as
can serve
heating and air conditioning systems (HVAC), where air quality monitoring
andlor
flow control and mixing of interior and outside air is of interest. Such HVAC
system
can be controlled by a central controller (e.g., controller 34), to meet user
needs
automatically or on demand.
DMS devices of the invention may incorporate various electrode
configurations, including coaxially or non-coaxially cylindrical, curved,
curvilinear,
arcuate, radial, plate, parallel, planar or flat. These configurations may be
focusing
or non-focusing as practiced in DMS devices. A preferred practice of the
present
invention is generally referred to as "plate-type", and it will be understood
embodiments may use facing electrode portions, segments, sections, or plates.
In one
embodiment, non-uniform focusing fields are generated between the DMS filter
electrodes; such embodiment may include a curved flow path including flat, non-
flat
or curved DMS electrodes.
Turning now to FIG. 4A and FIG. 4B, alternative embodiments of electrodes
20 and 22 are shown. As shown in FIG. 4A, filter electrodes, labeled 20' and
22',
can be coaxially cylindrical. In another embodiment, shown in FIG. 4B, either
one
or both electrodes, labeled 20" and/or 22", can be curved or curvilinear. In
particular, segments 20A can be curved, thereby producing a focusing effect
known
in the art, or straight, they eby producing a field substantially similar to
that produced
by the electrodes 20 and 22 as showwnn in FIG. 2B. In the embodiment where
segments 20A are straight, their length can be variable..
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-40-
DEVICE FOR DETECTION OF ANALYTES IN A VOLATILIZED SAMPLE
In one embodiment, the present invention is an apparatus for detection and
identification of allalytes in a volatilized sample using the mobility-based
signature
obtained by a differential ion mobility spectrometry (DMS) device.
Deferring to FICa. 5, an illustrative pyrolysis-based DMS system device of
the invention will now be described. It is understood that any of the
volatilization
techniques described herein can be adapted for use with the instant
invention..
In this embodiment, system 100 includes a sampler or other pal-ticle collector
102 which delivers liquid or solid sample to a pyrolyzer 104. (such as a
commercially available pyrolyzer from CDS Analytical) which has an output
coupled to the flow path 106 of the DMS analyzer 110. The flow path structure
(sometime referred to as a drift tube) has an inlet 112 for receipt of the
pyrolyzed
sample output from the pyrolyzer carried by carrier gas 114. The pyrolysate in
transferred from the pyrolyzer to the DMS through a sealed and heated
interface.
During sample loading on the probe, the pyrolysis chamber is purged while a
stream
of clean N2 is diverted into the DMS. During pyrolysis, the flours are
diverted
through a valve into the analyzer to assist introduction of the pyrolate.
In one example, the pyrolyzer heated samples from room temperature to
1400 C at rates from 10 - 20 °C lmsec. The controlled temperature
ramping enables
selective desorption of compounds from the probe, therefore enhancing
resolution
and signal-to-noise of the apparatus. A drying function evaporates and vents
the
solvent out a purge vent resulting in sample concentration and prevention of
the
solvent from entering the DMS analyzer 110. A probe cleaning function, flash-
heats
and desorbs left-over sample between analyses.
The pyrolate is carried by the carrier gas into the ionization chamber 120
where source 122 ionizes the sample. The ions ("+", "-") are carried by the
carrier
gas into the filter 124 between filter electrodes 126, 128. In a preferred
practice of
the IllVelltlOn, an asymmetric high-low varying RF field is generated between
the
filter electrodes, with applied DC compensation under control of
controller/dhiver
130. Ton species are passed to the detector 132 based on compensated field
conditions and mobility difference f~r the species. As the compensation is
scanned,
a spectrum can be recorded for the sample. Detector 132 includes electrodes
134,
136, which enable detection of positive and negative modes fox each species.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-41-
In a preferred embodiment, electrodes 124, 134 are formed on a substrate
140 and face electrodes 126, 136 formed on substrate 142. Preferably the
substrates
are insulating. The substrates are mated to fix the distance between the
electrodes
aazd defining the analytical gap G betvJeen the electrodes (preferably
~ODSmxn). The
asyrrnnetric field is generated between these electrodes transverse to the
analytical
gap and the ions are flowed in the gap through the field.
~DITI~NI~L ILLUSTRATIVE ~IIZBQI~II~ftENT ~F DIES PRACTICES OF THE PRESENT
w.
INVENTI~N
Deferring now to FIG. 6, an alternative separation method and apparatus of
the invention is shown. In this embodiment, a dual channel system 300 includes
a
first flow path 11 and a second flow path 311. As the ions are separated by
passing
through filter 24 in flow path 11, they are deflected by deflector electrode
320 into a
detector 326. The deflector electrode 320 is disposed substantially at the
output of a
flow path (here, a first flow path 11) of a differential mobility spectrometry
(DMS)
device. Preferably the detector may include a detector electrode 327 which
also can
be biased to act as an attractor electrode 327 in combination with the
deflector 320.
In a non-limiting example of the operation of this embodiment, positive ions
(T+)
will be deflected by a negatively biased deflector 320 and positively biased
attractor
327, while a negative ion (T-) will either continue along its original path or
will be
neutralized on a negatively charged deflector electrode 320. In one
embodiment, as
the ions strike the detector electrode 327 their charges are registered and
generate a
detection signal. Now detection spectra can be generated as needed. In the
embodiment of FIG. 6, the ions are separated from the rest of the flow out of
the
filter 24 in flow path 11, and by mea~ZS of the deflector, a refined set of
ion species
of interest is detected and/or obtained. As this species is neutralized by
contact with
the detector electrode they pass out of the detector/collector as a purified
set of
molecules which may be used as a collected sample, may be re-ionized and
reprocessed, or the like.
It will be further appreciated that in embodiments of the invention,
detections
are made and then the identification process typically involves comparison
against a
loolcup table of stored detection data. Thus a practice of the invention not
only
results in detection of a marker but also results in indication of the analyte
with
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-42-
which the marker is associated. For example, if bacterial spores were in a
sample,
the above detection results would be obtained and would be compared against a
store ofrelated detection data. Upon a positive match, an identification
announcement would be made.
Preferably, the apparatus of the invention includes an on-board volatilizer
portion and DMS analyzer, wherein collected samples axe volatilized and then
resulting gas sample is automatically transported to the analyzer and then
detected
for evaluation of presence of analytes in the sample based on ion mobility
signatures. W a further embodiment, the volatilizer and DMS device may be made
in
a single package. The sample collector may also be on-board.
In a further embodiment of the present invention, a sample is identified by a
mufti-stage analysis, wherein a first stage filters a sample by particle size
and
defines a narrowed sample set, and in a second stage this sample is pyrolyzed
and
then analyzed based on high field ion mobility as discussed. Results of the
first and
second stage axe correlated with known standards to identify the compounds in
the
sample.
MULTICHANNEL ~ARRA1'~ DETECTION
Devices suitable for practicing the methods of the present invention are
described in U.S. Fat. No. 6,495,23 and U.S. Pat. No. 6,651,224, and include
teaching an array of DMS filters. .An illustrative device comprises a housing
defining at least one flow path between a sample inlet and an outlet, a
plurality of
ion filters disposed within the housing, each ion filter including a pair
spaced filter
electrodes, and an electrical controller for applying a bias voltage and an
asymmetric
periodic voltage across each pair of ion filter electrodes for controlling the
path of
ions through each filter. In one embodiment, the device provides an array of
filters,
each filter associated with a different bias voltage, the filter may be used
to detect
multiple selected ions without sweeping the bias voltage or, in an alternative
embodiment, by simultaneously and independently sweeping the bias voltage in
different ranges and at different fields. Filters may be in parallel or in
series with one
chemical sample processing through multiple ion filters.
The teaching of the above-referenced disclosures are incorporated herein by
reference in their entirety.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
- 43 -
EXEMPLIFICATION
E~~I~t~r.E 1
ENDOSF~RE EI~IvI~I~EER 2,6- P~'RIDINE-DICARE~~YLIC A,C~ (DIPIC~L1NIC ACla) IS
DETECTED BY ~ DMS DEVICE AND ~~ICILLfIS SP~RE BI~Mf~RI~ERS ARE DETECTED EY h
DMS L)EVICE AFTER PYR~LYSIS.
Pyrolysis of bacterial spores from species such as ~acilla~s and
Cl~str°idiurra
produces large quantities of gaseous, 2,6-pyridine-dicarboxylie acid
(dipicolinic acid
or DPA) as wzique marker, which may then be detected by a gas
chromatographylDMS device. Typical spores contain roughly 5-15% dry weight of
DPA (MW = 167), which is speculated to provide the spore with heat resistance.
While the presence of DPA does riot signify with certainty that an infectious
agent is
present in the environment, a sudden increase in its concentration can serve
as a
trigger for initiation of a target-specific search.
The conunercially available pyr~lyzer PyroPrabeI000 eras acquired from
CDS Analytical with the necessary functions to handle the introduction of
liquid and
solid samples into the DMS detector. The pyrolyzer is capable of heating
samples
from room temperature to 1400°C at rates from 1 to 20°C/s. The
controlled
temperature ramping enables selective desorption of compounds from the probe,
therefore enhancing resolution and signal-to-noise of the DMS. A drying
function
evaporates and vents the solvent out a purge vent resulting in sample
concentration
and prevention of the solvent from entering the DMS filter. A probe cleaning
function, flash heats and desorbs left-over sample between analyses. The
pyrolate is
transferred to the DMS through a sealed and heated interface. During sample
loading
on the probe, the pyrolysis chamber is purged while a stream of clean N2 is
diverted
into the DMS. During pyrolysis, the flows are diverted through a 6-port valve
into
the DMS f~r intr~duction of the pyrolate into the DMS.
In order to provide a control result for comparison with data obtained using
the DMS unit, a ~. subtili,s sample was analyzed using pyrolysis-ion trap mass
spectrometry. The concentration of the ~. S~cbtiliS sample was 10'
organisms/ml.
This experiment (FIG. 7) showed that the expected biomarkers and both DPA and
picolinic acid (PA) were evident in the endospore spectra. Other unidentified
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-44-
biomarkers were also measured using this technique as shown in the background
spectral peaks. We further tested the ability of the DMS to detect endospore
biomarkers by using both DPA and PA standard solutions.
FIGS. ~ (dual mode), ~ (dual anode) and 10 (single mode) provides spectra for
PA and DPA obtained fxom solid samples pyrolyzed sequentially and detected in
a
DMS system of the invention. Picolinic acid was pyrolyzed through a
temperature
excursion of 130 to 300°C at a rate of 20,000°C/s, the interface
temperature was
held at 130°C. Dipicolinic acid was pyrolyzed from 145 to 400 °C
at 20,000 °Cls,
the interface temperature was held at 145 °C. both PA (100 ppnl) and
DPA (100
ppm) produce positive and negative ion peaks that can be used for
identification. In
addition, pure DPA produces a secondary positive ion peak, further
differentiating
its fingerprint pattern. The peak width at half height averages 1.4 V. It is
known that
pyroLysis is capable of fully decarboxylating DPA to pyridine. Ideally,
controlled
and more gradual' pyrolysis conditions will lead to loss of only one
carboxylic acid
group to generate PA, enabling specific identification of the DPA source as
bacterial
spores. Due to the volatility of pyridine, pyrolysis was not necessary for
introduction, and the interface temperature was held at 130°C. As seen
in the DMS
spectra, pyridine does not produce negative ions. The absence of a negative
ion peals
enables one to conclude that the pyrolysis conditions employed are mild enough
to
prevent full decarboxylation and that pyridine can be differentially detected.
FIG. 11A shows the full DMS spectrum of pyrolyzed B. subtilis spores as a
simulant for B. ah.thracis. The spores were suspended in dH2~ at a
concentration of
2.4 x 109 spores/ml. The spores were diluted to 100,000 spores in 1 ~,1, and
the
sample was dried and pyrolyzed tlarough a temperature excursion from
250°C to
400°C at 20,000°C/sec with the interface temperature at
250°C. FIG. 11A shows
the time course of the DMS spectra after pyrolysis is initiated. The signal is
complex
and changes over time, indicating a large number of potential biomarker
targets.
There is a large prominent peak as well as low amplitude clusters of "noise"
in the
spectra. The prominent peals array be a complexed forn ~ of DPA or PA released
during pyrolysis, although exact chemical identification is not possible. The
DMS
signal may also contain two components in the low amplitude signal: electronic
noise, and trace levels of organic volatile compounds. FIG. 11B shows a time
course
of both the positive and negative DMS spectra at 10 seconds after the onset of
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
- 45 -
pyrolysis. Both spectra show biomarkers are detected. These results show the
DMS
is capable of detection of known endospore biomarkers.
E~I~PLE 2
M~NIT~R1NG F~~D QUALITY
Chemical changes an the living system or degradation processes of cells after
death are accompanied by the formation of molecular byproducts. These
processes
include the breaking down of peptides and DNP~ strands to smaller components,
and
changes in amino acids that lead to the formation of amines. ~ne of the
processes of
particular interest is the breakdown of amino acids and the production of
diamines
and polyamines.
Furthermore it is known that bacterial decarboxylation of ornithine and
lysine produces putrescine and cadaverine respectively. An atmospheric
pressure
ionization method of the invention is particularly suited for the detection,
of these
markers, such as biogenic amines, since they tend to have either high proton
affinity
and form stable positive ions or high electro-negativity and readily form
negative
ions that are detected in a DMS system of the invention.
FIG. 12 shows DMS spectra detected according to a food-qualzty
embodiment of the invention for a mixture containing both putrescine and
cadaverine. The putrescine peak at about -30 volt compensation is well
separated
from the cadaverine peak at -29 volts, and which are separate from the
detected n-
Nonylamine. Based on these results, it will be appreciated that a food-quality
detector of the invention can be used to evaluate the quality of a food
sample, such
as meat, based on the detected presence and intensity of these bio-markers.
EXAMPLE 3
BREATH l~NALYSIS
Another application of the invention is in breath analysis. The hiunan breath
contains over 4.00 organic compounds at concentrations typically in the parts-
per-
million (ppm) to parts-per-billion (ppb) range. ~nly a slender barner, the
puhnonary
alveolar membrane, separates the air in the alveoli of the lung from the blood
flowing in the capillaries. This membrane allows volatile organic compounds to
easily diffuse from the blood into the breath. Moreover, the concentration of
these
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-46-
compounds in the breath can be correlated to their concentration in the blood,
as
noted through the widespread use and acceptance of a breath analyzer to
determine
alcohol consumption.
Through systematic studies, concentrations of pauticular compounds have
been correlated with specific diseases or impairments in metabolic pathways.
However, while these studies are encouraging, there are still a n~.unber of
complicating factors which have limited wide spread adoption of breath
analysis for
medical diagnosis. These include: the complexity of current breath analysis
systems,
their high cost, amount of correlation bett~aeen the data and disease, and the
complexity of data analysis due to interferences and moisture.
In practice of the present invention, a non-invasive DMS breath analysis
system is provided. W one set of experiments, sample collection involved
collecting
a breath sample directly onto a solid phase micro-extraction (SPME) fiber
assembly.
The SPME fiber was placed in proximity to the mouth of the subject and the
sample
collected for two minutes. The SPME assembly was coupled to a GC injector port
which was held at 120C and desorbed the sample from the fiber into the GC
column.
The present wide-spectrum DMS was attached at the detector port of the GC for
DMS filtering and species identification of the GC elute.
A background baseline spectra without sample on the SPME fiber is shown
in FIG. 13. Spectra from subject #l, FIG. 14, and subject #2, F1G. 15, are
very
similar except for the peak at a compensation of about -3 volts for specimen
#2.
Using the GC alone, without the benefit of the present invention, the presence
of
these different compounds would not be evident. The resultant GC-DMS plots
shows the chromatographic retention time on the y-axis and the compensation
voltage plotted on the x-axis and shows the value of the detector in providing
additional information to simplify and assist in the analysis of a human
breath
sample, as a viable practice of the present invention. However it also should
be
noted that in practices of the invezation direct sampling and analysis by DMS
can be
practiced without SPME.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
_q,7_
Example 4
B. SUBTILIS SPORES DETECTION
As shown in FIG. 16, spectra for markers from pyrolyzed B. s~cbtalas were
identifieda Spectral DMS scans for pyrolyzed water sample are shown in A~
4.0,000
spores pyrolyzed are shown in B9 and 120,000 spores pyrolyzed are shown in C.
A
person spilled in the art will recognize from this data that marlcers at 1, 3
and 4
correlate with the pxesence of spores, with amplitude corresponding to
concentration.
EXAIvIPLE 5
ANALYSIS OF ML1RINE URINE SAMPLES
In one experiment mouse urine samples were tested using DMS as shown in
FIG. 17 showing positive DMS spectra. In the figure, a large carrier gas (N2)
peak
(~ 3 a.u.) is seen at 0 Voltage Compensation (Vc), while urine headspace
(vapor)
detection spectra is seen just to the left (< 0 Vc).
hi tlus demonstration, urine sample headspace vapor from three different
individual B6-H-2b male mice was analyzed. The DMS spectra indicated the urine
samples were similar to each other but different from two control
monomolecular
odorants (isovaleric acid and isoamylacetate). A small number of sample
preparation
permutations were tested to identify conditions that yielded the most
volatiles as
indicated by intensity in the DMS spectra. Addition of (1) water (to increase
volume of urine sample), (2) salt (0.2 mg/~,1), and (3) heat (37 C), all
yielded more
volatiles.
It was possible to demonstrate use of DMS as a biological evaluation tool for
medical diagnostics, such as for urinalysis. It is also noted that such
testing does not
require fresh liquid samples. While freezing and thawing of such samples
reduces
the amount of volatiles, still detection can proceed. In practice of the
invention, a
data base of urine and analyte samples can be determined and stored.
Further, a lookup function of a device of the present invention enables
identification of ion species detected in urine. This experiment is
illustrative of
establishing baselina upon which a specific detector can be established. For
example, detection of such analytes as pronase, (NH4)aS04, KHaF04-H20, CI~2O3,
KaCO3 or NaCI can be detected.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
_q.8_
EXAMPLE 6
USE OF A DMS DEVICE OF THE PRESENT INVENTION AS A CHROMATOGRAPHIC
DETECTOR
A DMS device suitable for practice of the present inven tion was interfaced t~
a GC and used as a chromatographic detector. The system performance was
compared with the Flame Ionisation Detector (F~). The average F~ detection
limit was 2E-lOg, while a preferred DMS system of the invention had a
detection
limit of 2E-11g. Furthermore, the DMS is flame-free.
Similarly to a mass spectrometer, the ion information provided by the
invention offers a second dimension of information to a GC chromatogram and
the
ability to enhance compound identification. FIG. 18 shows spectra according to
a
GC-DMS embodiment of the invention, with the part shown as a chromatogram
(right frame) being typical of what is seen from a FID. In practice of the
invention,
the chromatogram is the sum of the peak intensities for the product ions
created. The
associated two-dimensional plot (left frame) of ion intensity (indicated by
gradient)
versus scanned compensation voltage provides a means of fingerprinting the
compounds eluted from the GC. Therefore practice of the invention provides
three
levels of information: retention time, compensation voltage, aald ion
intensity, all
shown on the spectra of FIG. 18. Furthermore, in a preferred in-line DMS
system of
the invention such as taught in US Patent No. 6,495,823, spectra may be
obtained
simultaneously for positive and negative ions, i.e., dual mode, augmenting or
eliminating the need of serial analysis under possibly changing instrumental
conditions, as required with other equipment.
As shown in FIG. 19, decreased GC runtime produced co-eluting species that
were subsequently resolved in the DMS spectra. In this way, a fast GC can be
used
while maintaining the required compound resolution. Furthermore, the
reproducibility of the present invention compares very well to that of the FID
as
shown in FIG. 20. FIG. 20 shows a c~mparison of FIL2 and DMS reproducibility
for
a homologous alcohol mixture.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-49-
EXAMPLE 7
PARALLEL ANALYSIS OF BACTERIAL SAMPLES USING PYROLYSIS/GAS
CHROMATOGRAPHY AND A DMS, FLAME IONIZATION AND MASS SPECTROMETRY
DETECTION hEYICES
S The py-GCIDMS analysis of bacteria showed a broad range of ~rolatile and
semi-volatile organic compounds spanning molecular weights from SO to over 250
amu. Information contained in the patterns of retention time versus
compensation
voltage prove analytical value of the differential mobility spectrometer.
Products
from the pyrolysis of bacteria were matched to lazown chemicals. The findings
were
also supported by parallel studies using pyGC/FID and py-GC/MS.
Material and Methods
Detecti~n Devices
Three gas chromatographs (Hewlett-Packard Co., Avondale, PA) were
1 S equipped with a splitless injector, 1 S m SPB-S capillary column (m 0.25
mm, 0.25
~,m filin thickness, Supelco, Bellefonte, PA) and different detectors. Each of
two
HP model S890A gas chromatographs was equipped with a flame ioiuzation
detector or an HP model 5871 mass selective detector (MS). An HP model 5880A
was equipped with a DMS analyzer as detector. Experimental parameters for all
gas
chromatographs were identical and included: initial temperature, SO °C;
initial time,
2 min; program rate, 8 °Clmin; final temperature, 2S0 °C; and
final time, S.SO min.
Pressure on the injector ports was nominally S psig with a split ratio of 50:1
and
was adjusted individually so retention times between instruments matched.
Split
flow was -30 mL/min and septum purge was 3 mL/min. Bottled nitrogen (99.99%)
2S was used as carrier gas for the GC/Fm and GC/DMS. Helicon (99.99%) was
scrubbed over a Hydrox Purifier Modal 8301 catalytic reactor (Matheson Gas
Products, Montgomeryville, PA, USA) and used as carrier gas for the GC/MS.
Parameters for the F1D with integrating recorder were: threshold, 3; area
reject,
100; and attenuation, 2. Parameters for the MS were: mass range, SO to SSO
amu;
threshold, 500; scan rate, I.S scans/s; and electron multiplier voltage, 1600
~
according to the automated calibration routine.
The differential mobility spectrometer contained a planar micro-scale drift
tube made from ceramic plates with gold plate copper base electrodes. The
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
separating electrodes were 4 mm wide x 6 mm long and a width of 0.52 mm was
used for the gap between the electrodes. Electrodes were fixed in a metal body
to
which a -1 mCl 63N1 lon source waS attached. The ion source was connected t~ a
trap afar line into the OC o~'en. Air at 0.5 L/min was heated and passed into
the
transfer line for gas flow through the DIMS analyzer. The transfer line and
DIES
analyzer were maintained at the same temperature. Air was provided using a
pure
air generator (Mode 737, Addeo Core, II~liami, FL) and was further purified
through
beds of 13x molecular sieve. laloisture in the carrier gas was monitored using
a
model MIS-2 meter (Panametrics, Inc., Waltham, MA) and was 30 ppm or below.
IO The drift tube was operated using in-house built electronics containing an
I~F
wavefonn generator, a sweeping voltage generator, and an electrometer. The
waveform generator was based on a soft-switched, semi-resonant circuit that
incozporated a fly-back transformer and allowed variable peak-to-peak
amplitudes
of the asymmetric waveform from 200 V to 1600 V without altering the waveform
shape. The operating frequency of the RF generator was 1.3 MHz. A compensation
voltage ramp was synchronized with the data collection system and provided a
scan
of compensation voltage from -10 V (or -5 V) to +30 V (or 24 V) at a frequency
of
1 scan every 3.8 (or 2.8) s. Signals were processed using a model 6024E
National
Instruments (Austin, TX) board; spectra were digitized and stored for every
scan
using software derived from Labview (National Instrurrlents).
The pyrolysis apparatus was model 150 Pyroprobe (CDS, Inc., Avondale,
PA) pyrolyzer with a platinum ribbon probe. The probe was housed during
analysis
in a glass chamber, with a gas flow inlet and needled outlet. Dimensions of
the glass
chamber were: inner diameter, 7.2 mm; outer diameter, 9.2 mm; and length, 15
cm.
The chamber was attached to the pyrolyzer through a 3/4" tube union (Swagelock
Company, Solon, OH) and the needle attached to the end of glass chamber via
114-
1/16" Swagelolc reducing unit. The total volume of the chamber exclusive of
the
probe was 3.6 mL. Nitrogen was provided to the chamber at 44 mL/min during the
r~.n, so the gas volume of the chamber was replaced every 5 s. The chamber was
wrapped with a resistive wire heater and insulated with glass pack to maintain
temperature at 250 °C during transfer of the pyrolysate to the CsC inj
action port.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-51-
Bacterial cultures ahd growth couditioh
Escherichia coli I~-12 (strain # 25), Micrococcus luteus (strain # 52) and
Bacillus uaegateriuyn (strain # 61) were obtained from the New Mexico State
University Culture Collection. The three cultures were gTOVm for 12 hrs in
nutrient
broth (Difco, Detroit, MI) in a~z orbital shaker (150 I~IVI) maintained at 30
°C. The
cultures were harvested by centrifugation (10,000 l~Il4, 2 min.), re-suspended
in
sterile water, spun again, and then re-suspended in sterile water. The cells
were
quantified by sequentially diluting the Bells and plating sub-samples onto
nutrient
agar plates, and counting resultant duplicate samples. Under these growth
conditions, 66 ~/o of the B. traegateriuTya cells had sporulated as measured
by staining
with Brilliant Green and microscopic examination. The cells were pyrolyzed and
analyzed on the day of preparation, or were stored at 4 °C and analyzed
within one
week. Biomarkers included Lipid A {CaibiochemNovabiochem Co., La Jolla, CA) ;
Lipoteichoic Acid (from BIOTREND Chemikalien Germany); and dipicolinic acid
(Aldrich Chemical Co, St. Louis, MO) and were used as received without further
treatment.
Procedures
Several procedures were used throughout the whole of this study and
included handling and injection of the sample from the pyroprobe apparatus to
the
GC, data collection using the GCfDMS, data reduction to spreadsheets, and
analysis
of the data.
Before each measurement, the pyroprobe containing the Pt ribbon was heated
to 800-900 °C or inserted into the flame of a Bunsen burner to remove
residues of
previous measurements. Sample volumes of approximately 20 mL were applied by
micropipette onto the surface of the Pt ribbon and dried at -75 °C in
air for 1 to 1.5
minutes. The temperature of the glass chamber of the pyroprobe was 250
°C.
10 lCL of bacterial sample was placed. on the pyroprobe ribbon. Directly
before the GC analysis, the pyrolysis apparatus with the pyroprobe containing
bacteria was purged with nitrogen at a gas flow of-44 mL/min for -6 s (one
replacement volume of the pyrolysis chamber). The nitrogen flow was stopped
and
the needle of the apparatus was inserted through the septum of the injection
port of
the gas chromatograph. Nitrogen was again applied to the pyrolysis apparatus
and
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-52-
the sample was pyrolyzed at 650 °C for 10 s. The gas chromatographic
analysis was
started simultaneously with the start of pyrolysis and the DMS analyzer was
operated continuously with mobility scans obtained every 2.8 s (cf. above).
After 30
ss the pyrolysis gas flow was stopped and the pyroprobe was removed from the
injection port. Since the widths of individual GC peaks were 5 to7.5 s at the
baseline
during an elution profile, two to three differential mobility spectra were
recorded for
each peak eluted in the chromatogram.
The positive and negative spectra of each py-GC/DMS run were saved as
ASC 11 files (file size 1.3-1.5 MB). These files were imported into ~rigin 6.1
and
plotted into graphs. Quantitative determinations were made using chromatograms
through Peakfit 4.0 (Tandel Scientific, San Rafael, CA). Plots of ion
intensity versus
retention time were deconvoluted with these PeakFit parameters: for automatic
baseline subtraction, Linear, 2%; peak smoothing, FFT Filtering in levels from
10 to
24%; Auto Place and fit peaks, Linear two point baseline; FFT Filter, 28 to
54.17
smoothing, options, Chromatography; and Gauss area, 1.5 % amplifier.
Results
Pr~files fi°of~a pyrolysis of bacteria with GCIDMS
Results from the pyrolysis of three bacteria with GC/DMS analysis are
shown as plots of ion intensity, retention time, and compensation voltage in
FIG. 21
and FIG. 22. In FIG. 21, plots for positive ions axe shown from the pyrolysis
of E.
coli, M. luteus, and B. megatef~ium in frame A, B, and C, respectively.
Constituents
were seen above background throughout a column temperature range from 50 to -
190 °C (retention times of 0 to 20 min) and between compensation
voltages of -2 V
to 8 V. The peaks in FIG. 21 arise from reactions (Eq. 1) between substances
in the
GC effluent and a reactant ion (Hk(H20)") seen at a compensation voltage of -
14 V
(not shown).
M + H'-(Hz~)" ~ MH+(H2~)" + H20 (1)
Molecule Reactant ions Production
In the differential mobility spectra, the more offset peaks (such as at higher
values of compensation voltage) will be of lower molecular weight than those
at 0 V
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-53-
and are consistent with molecular masses of -50 to -124 amu. Peaks with
compensation voltages of 0 to -3 V are found with ions off high mass which
exhibit
yaegative dependence of mobility on electric field strength (approximately 150
amu
and higher). Thus, the compensation voltage a~~is can be viewed as a n Measure
of low
(at ~ V) to high (at -3 V) mass spanning a range from approximately -50 to 250
amu.
Each bacterium produced a pyrolysate with a eomplex mixture of volatile
compounds and this was observed in results from the GC/DMS, GC/F'1D and
GC/MS. In FIG. 21, a general trend can be seen in the plots for increases in
molecular weight with increases in retention time. However, some differential
mobility spectra can be seen with two peaks at differing compensation
voltages.
Commonly in IMS, a protonated monomer will form a cluster with a sample
neutral
to form a proton bound dimer, M2H+(H20)" , when sample vapor concentration is
increased. The proton-bound dimer will appear at the same retention time as
the
protonated monomer though compensation voltage will be displaced in the
direction
of zero that for the protonated monomer. This can be an additional component
in the
orthogonal differential mobility spectrum as a measure of concentration or
abundance. An example of this is evident in FIG. 21 C with a component at a
retention time of 6 min. where two peaks are seen at 3.4 V and -1 V in the
differential mobility spectrum. These peaks axose from the same substance,
crotonic
acid, and were the protonated monomer and proton bound dimer. In the
corresponding GCIMS data set, a single chromatographic peak was observed with
the mass spectrum of crotonic acid orthogonal to retention time. Tlus can be
seen
through the plots for other substances. 'The complexity of the plots from py-
GC/DMS is consistent with findings with py-GC/FID and py-GC/MS analyses made
using the same samples under identical conditions of pyrolysis. The
chromatograms
could be matched among all three instruments within each of the three bacteria
samples. The number of constituents, resolved and detected, was 50 to 70 using
py-
GC/F~, py-GC/DMS or py-GC/MS for the total ion chromatograms.
The reproducibility of the py-GC/1~MS plots was determined for
compensation voltage, peak intensity and retention time and average values for
these
were X0.2 V, 10 °/~ relative standard deviation (RSD), and X0.05 min,
respectively.
The GC/DMS alone with chemical standards injected by syringe yielded
reproducibility of X0.1 V, 10% relative standard deviation (RSD), and X0.02
min for
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-54-
compensation voltage, peak intensity and retention time, respectively. This
demonstrated that the contribution to variance from sample handling by
pyrolysis
was negligible when compared to the same for a syringe injection, noted above.
Moreover, comparable quantitative variance wa.s obtained from py-GC/F~
analyses
suggesting that the I~MS as detector was not intTOducing significant variance
into
the analyses versus the FIFJ. Consequently, any differences seen in the
patterns of
FIG. 21 are not variability of the I~MS but can be associated with chemical
differences of the bacteria as discussed below.
In a micro-fabricated I~MS analyzer, vapors ionized in the source region are
swept into the analyser region where positive and negative ions are pushed
through
the drift tube and characterized simultaneously. Results for negative ions are
shown
in FIG. 22 and came from the same data set shown in FIG. 21. The chemistry of
ionization is based upon reactions between substances in the GC effluent and
negative reactant ions, here, 02 (H20)". which was evident at 11.3 V (off
scale in
FIG. 22), (see Eq. (2)).
M+ O2- (HZO)" --~ MOZ (H20)n 1 + HZO (2)
Sample Reactant ions Product Ion
Ionization chemistry with negative ions is generally more selective than that
for positive ions and based upon 02' (H20)" attachment to a neutral. This
occurs with
molecules containing acidic protons or electronegative groups. In some
instances,
the product or adduct ion (M02 (H20)"_1) may dissociate to form M- or M-H-.
The
plots of retention time versus compensation voltage for negative ions also
showed a
large number of constituents between 0 to 20 min elution times and
compensation
voltages from 10 to 0 V. However, fewer constituents were observed for
negative
ions versus that for positive ions with from 31 to 39 peaks detected and
resolved.
This is consistent with the anticipated increase in selectivity for negative
ionization
cheix~istry. Crotonic acid was noticeable in the negative ion py-(~CJI~l~IS
plots and
was expected since carboxylic acids exhibit favorable ionization chemistry
with OZ
(H20j".
The availability of information for negative product ions is an additional and
separate measure of chemical identity over positive ion response and is
available
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-55-
conveniently with a py-GC/DMS measurement. Such chemical information might be
correlated to the response with positive ion chemistry. Alternatively, the
capability
for response with negative ions is an opportunity to employ derivatizing
agents that
are particularly well suited for negative ions (fluoro or halo derivatizing
agents that
have been used with ECI~s~.
The results in FIG. 21 and FIG. 22 were compared favorably to results by
py-GC/F~ and py-GC/MS with the same sample. The results in FIG. 21 and FIG.
22 demonstrate that a 1~I~1S analyzer shows response with detail for molecules
spanning a range ~f molecular weights from 50 to 250 amu (gauged from
retention
time or temperature) and that the existing I)MS analyzer provided resolution
of
chemical information orthogonal to retention.
~'y-GCIDiVIS analyses of biapolyfners
Biopolymers for some major constituents in bacteria are available as purified
substances though the choice of materials is limited by genus and species.
Lipid A
and lipoteichoic acid were obtained commercially anti mere characterized by py
GC/DMS in order to compare these prospective sources of bacterial chemical
information to actual results from bacteria samples. Studies were made of the
biopolymers alone and of biopolymers as mixtures with bacteria. Negative
controls
were made by mixing biopolymers with bacteria missing the biopolymer. Results
from these studies are shown in part in FIG. 23A to C from py-GClDMS
determinations of the Gram-negative bacterium E. coli, Lipid A, and a mixture
of E.
coli with Lipid A, respectively. The findings show that some of the peaks from
E.
coli can be matched to peaks from Lipid A using retention time and
compensation
voltage. However, peaks can be seen in the plot for E. coli which are not
associated
with and in plots from analysis of Lipid A and peaks can be seen from Lipid A
that
are not evident in E. coli. The positive control with a mixture showed that
differences were not due to matrix effects of any chromatographic
uncertainties such
as adsorption at active sites. Rather, the profiles were additive as shown in
FIG. 23C
and the differences cannot be attributed to pyrolysis, GC separation, or
ionization
chemistry in I~1~J1-S analysis. These differences are likely due to the ~rigin
and
composition of Lipid A which came from Salmonella minnesota and not from the
genus Escherichia. Thus, the patterns seen in the plots for E. coli may be
understood
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-56-
to arise from pyrolysis of biopolymers other than Lipid A. Results from py-
GC/DMS analysis of M. luteus, lipoteichoic acid, and a mixture of M. luteus
with
lipoteichoic acid showed different patterns from those in FIG. 23, though the
virtually identical conclusions as that with Lipid A. In this instance, the
lipoteichoic
acid was isolated from ~'tf-elat~coecus ~yi ogei2es. Consequently,
associations
between peaks in bacteria with pyrolysis products from biopolymers were
unsuccessful. Instead, chemicals known as pyrolysis products from bacteria
were
obtained as authentic chemical standards and were used to evaluate results
from py-
GC/I~MS and py-GCIMS.
GCl1?ll~f.S cl2aracte~izatioya of authentic standards, f'or claemicals front
~y~~olysis of
bacteria
Volatile and semi-volatile organic compounds are produced from the
pyrolysis of bacteria and these have been recently described by Snyder et al.
Samples of most of these chemicals were obtained connnercially as authentic
standards and were characterized for retention time (tr) and compensation
voltage
(C~) by GC/DMS, and for retention time and mass spectra by GC/MS. All
chemicals
except some carboxylic acids exhibited distinct chromatographic retention,
distinctive compensation voltages, and mass spectra which matched reference
spectra (exceptions included carboxylic acids which, apart from a few
exceptions,
were either adsorbed on active sites in the injector port or showed poor
chromatographic efficiency on the non-polar stationary phase). The location of
these
chemical standards in plots of retention time versus compensation voltage are
shown
in FIG. 24 and FIG. 25 with numbers overlaying points for abundant peaks from
py-
GC/DMS analysis of E. coli (70 peaks) and M. luteus (50 pealcs). Numbers,
referenced to the caption in FIG. 24, axe placed at the intersection of tr and
C" so
comparisons can be made, within the error of measurement, between authentic
chemical standard values to values of peaks created from pyrolysis.
Results of both chemical standards and two representatives of Gram-positive
and Gram-negative bacteria seen in FIG. 24 where comparisons shows that not
all
chemicals reported by Snyder could be found in plots for positive ions.
However, ~
of 16 chemicals possible were observed as matches in both requirements, i. e.,
retention time and compensation voltage, and improvement are expected as
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
_ S7 -
chromatographic conditions are optimized. This favorable comparison was
confirmed and supported by GC/MS analysis which demonstrated that the GC/DMS
provided chemical information consistent with the reference method (GC/MS) and
consistent with known pyrolysis chemistry described by others. The results
also
demonstrate that more than 75 °~° of the most ab~.mdance
chemicals in the py-
GC/DII~IS analyses are unknown. The importance of these other peaks in
disclosing
chemical information about bacteria is not lcnown and must be established in
further
studies.
Matches between authentic chemical standards and py-GC/DMS of bacteria
(30 peaks for each bacterium) with negative product ions were also evident in
a few
instances. As seen in FIG. 25, three of the chemicals gave favorable matches
with
bacterial plots and the remaining peaks were of unknown identity. These three
chemicals were also the only authentic standards to give negative ion
response.
These findings demonstrated that py-GC/DMS analyses provided some chemical
information that should be expected from bacteria.
Disc~~imihatiou betweeh Bacteria Usitzg Results from py-GC-DMS.
The central question in these studies was the suitability for py-GCIDMS to
provide analytical information to allow the discrimination between bacteria as
Gram-negative, Gram-positive, and spore forms. A particular interest was if
useful
information was encoded in peaks of strong intensity and hence ease of
comparison.
Results with spores were dramatic as seen in FIGs. 21 a~zd 22 where a
distinguishing
and reproducible peals or substance was formed uniquely with spores. This
chemical
was identified as crotonic acid using mass spectra, retention time and
compensation
voltage with an authentic standard. Though crotonic acid has not been used
previously as a biomarker for spores and is regarded as a chemical for Gram-
positive
bacteria in general, crotonic acid was not seen above detectable levels in
analyses by
py-GClMS or py-GCIDMS of M: ~ut~us. Rather, B. m~gater~ium with high spore
content only produced crotonic acid here. Tn prior studies, spores were
distinguished
by the presence of picolinic acid and pyridine or methyl derivatives of
picolinic acid.
Unfortunately, picolinic acid was absorbed on active sites of the injection
port when
solutions of 10-100 ng/~,1 were analyzed. Discrimination between Gram-negative
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-S8-
and Gram-positive bacteria required detailed attention to plots in FIG. 24 and
FIG.
25 and the identification of peaks unique to each bacterium type.
~f the 70 peaks seen in FIG. 24 from ~°. c~li, the majority were found
in
locations of retention time and compensation voltage distb~ct from ~~:
lacteacs. i~~hich
of these peals might be useful, alone or in combination, as biomarkers for
Gram-
negative bacteria and which will be too dependent upon cell history to be
analytically useful has not been determined. The large number and separation
for
other peaks is promising. Less promising is the distinctiveness ~f analytical
data for
Gram-positive representative, llrl. luteus. Few intense peaks were observed
for py-
GC/DMS analysis of ll~f luteus and most of these are coincident with peaks
from E.
coli as seen in FIG. 24. Only four peaks in the present instrumentation and
pyrolysis
methods can be distinguished from Grain~negative bacteria and these are seen
at
retention times of 5, 5.2 and 13.5 min. However, the unpromising condition is
substantially altered with chemical information is introduced from negative
ions.
Plots for negative ions are shown in FTG. 25 and provide another dimension of
chemical information (tr, C" and ion polarity) and in this instance, nearly
nine peaks
were observed for Gram-positive M. luteus and were thought to be
characteristic
markers for M. lutezcs. These are seen at retention times (min) of 2, 3.6, 4,
6.5, 9.5,
11, and 12.5. This demonstrates an advantage of DMS over traditional TMS with
the
simultaneous characterization of positive and negative ions.
Resolution and Sensitivity with Separation Field and Tetnpef~atur~e of the DMS
Ahalyzey~
In one practice of the invention, the separation voltage was varied to explore
effect on resolution between peaks on the compensation voltage axis. Results
are
shown in FIG. 26 from py-GCIDMS characterization of B. fnegate~imn for
positive
ions from four settings of the separation voltage (low to high, bottom to
top). In FIG.
26D, the reactant ion peak (Equation 1) is evident at compensation voltages
from 3
V to 5 V with a center at 4 V. Throughout this analysis, this peak is visible
and
exhibits a small drift in compensation voltage. This is attributed to a small
increase
in temperature of the daft tube during the GC column temperature program
caused
by poor thermal control of the drift tube. At low separation voltage, product
ions
appear throughout the chromatogram at compensation voltages between 2.5 to 0
V, a
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-59-
small band for distribution of analytical information. As voltage is increased
from
688 V to 860 V (FIG. 26C), the reactant ion peak is shifted from -4 V to -7.5
V
consistent with a positive alpha f~.tz~ction known for hydrated pxotons.
Product ions
in positive polarity also show cha ges in compensation voltage as the
separation
field is increased and some ions have been shifted to compensation voltages
higher
than those in FIG. 26A. These are known to occur through changes in AID and
are
characteristic of ions with masses below 150-200 amu, namely protonated
monomers c~f small molecules as marked in FIG. 26. ~ther peaks were shifted
toward a zero or negative voltage as the separation voltage is increased and
such
ions have negative dependence of mobility on electric field. In prior studies,
these
ions exhibited masses larger than 250 amu and have been associated with proton
bound dimexs. Thus, the presence of two peaks of differing compensation
voltage at
the same retention time can arise from distribution of charge between
protonated
monomer and proton bound dimer as governed by vapor concentration as found in
conventional mobility spectrometers. The advantage of this is that
concentration
information is available directed in the differential mobility spectra and the
range of
separation voltages allows control of resolution within boundaries.
Further increases in separation voltage to 944 V (FIG. 26B) and to 1032 V
(FIG. 26A) Iead to exaggeration of the patterns seen in FIG. 26D and FIG. 26C.
The
reactant ion peals is displaced to 14 V and a dramatic decrease in peak
intensity was
observed. This Ioss in intensity is observed in general with this DMS design
and the
cause is not fully described. A similax shift in compensation voltage and loss
in peak
intensity is also evident with product ions. The shift in compensation voltage
means
that resolution within the differential mobility spectra increases
considerably with
each increase in separation voltage; this is expected from plots of mobility
dependence with electric field. Nonetheless, a compromise between resolution
and
peak intensity was found at 944 V and was used throughout all the studies
presented
above.
Measurements fox negative ions paralleled the trends seen with positive ions.
Product ions were seen between -4 to 2 V at low separation voltage and
resolution of
peaks improved on the compensation voltage axis as the separation field was
increased. Unlike positive ions, negative ion clusters such as M20Z- are not
commonly observed at these vapor levels and the pattern with increased
separation
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-60-
voltage was comparatively simple with single ions for each chromatographic
peak.
Nonetheless, product ions were shifted toward greater AK and larger
compensation
voltages with increased separation voltage. Loss in peak abundance was
observed
also with negative ions with greatest loss occuruing bet~,een X44 and 1032 V,
consistent with the product ion behavior.
Detectors in gas chromatography are operated generally at temperatures 50
°C or higher above the maximum temperature applied to the column and
these
guidelines are intended to prevent sample condensation in the detector.
However,
temperature has secondary effects in IMS analysers including ion declustering,
"~ dissociation' or decomposition. Thus, a concern was the compromise between
sample condensation in the detector and distortion of analytical information
through
changes in ion stability with temperature increases. Results from py-GC/DMS
screening of B. rn.egate~ium with four temperatures of the DMS drift tube are
shown
in FIG. 27; gas temperature was determined in the flow vented from the drift
tube.
The pattern of peaks at two or more compensation voltages seen at 55 °C
(FIG. 27D)
are characteristic of the presence of protonated monomers and proton-bound
diners
routinely observed in mobility spectrometers including DMS at low
temperatures.
An example of this is the pattern at 6 min. where peaks appear at compensation
voltages of 3.5 V and 0 V for protonated monomer and proton bound diner,
respectively. However, as temperature is increased in steps, intensity for the
peak of
the proton bound diner declines and is missing at 115 °C (FIG. 27A).
Though a
slight decline in intensity for all peaks was observed with this temperature
change,
the pealcs at compensation voltages of I to -1 V, understood to be cluster
ions, were
lost from the DMS scans uniformly at all retention times.
Though decreases in temperature may be expected to alter the
chromatographic pattern through losses of sample by condensation for
substances
that elute at high column temperatures, there is no apparent loss of
chromatographic
detail above retention times of 10 min (115 °C) or 15 min (150
°C) when the DMS
is low temperature of 55 °C. At temperatures up to 110 °C, there
was no observable
increase in number of peaks eluted from the column though apparent sensitivity
of
the Dl~/iS decreased with increased temperature. After some weeks of py-GC/DMS
measurements with the DMS at 55 °C, decreased intensity in response was
observed
uniformly for all substances throughout the retention time scale and was
attributed to
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-61-
a gradual accumulation of sample as condensate in the drift tube. This
diminished
response was reversed by heating the drift tube to 110 °C or higher;
heating was
accompaaued by loss of reactant ion peaks which were replaced by a single peak
near 0 ~T. This was understood as excessive vapor levels of the source region
produced from off gassing of condensed sample in the source region. After some
hours the analyzer mss restored in clean response as seen in the differential
mobility
spectrum.
Comparable trends noted above for B, me~czterium in resolution, spectral
profiles and contamination at various temperatures were obtained also for E.
coli and
1V~ luteus. h. temperature for the DMS was a compromise between eventual loss
of
response at low temperature through accumulation or condensation of impurities
and
the loss of detail in differential mobility spectra from ion declustering at
high
temperature. A temperature of -90 °C was chosen in further studies and
provided a
balance between stable sensitivity over long periods and differential mobility
spectra
with multiple peaks or bands.
Quantitative Py-f~CIDMS of Bacteria
Mobility spectrometers equipped with radioactive ionization sources such as
63Ni exhibit proportional and quantitative response toward vapor
concentrations for
peak intensities in mobility spectra though linear ranges may be only 10-100.
Differential mobility spectrometers show detection limits from 10-100 pg for
volatile organic compounds and linear ranges of 100-1000. Response from py-
GC/DMS analysis of bacteria is shown in FIG. 2S for integrated peak area
versus
number of bacteria applied to the pyrolysis probe. The plots were made using
only
the axes for a single biomarker and the DMS was operated for maximum ion
resolution (a large separation voltage). Consequently, the plots in FIG. 2~ do
not
represent optimum conditions of temperature, separation field, or data
processing to
establish a limit of detection. Father, these studies were made to establish
if a
quantitative basis existed between the pyrolysis step and the observed
response. The
plots suggest that the sum of aII parts of the measurement including pyrolysis
and
DMS analysis are linear in the range explored. No additional efforts were
given to
processing the data through sum of all product ion intensity. Presumably,
additional
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-62-
dynamic range or improved detection limit should be possible if the signal was
processed and integrated or if only several biomarkers were integrated and
summed.
The minimum number of bacteria seen in this early studies was I O°
bacteria
where each bacteria showed characteristic response v~%here n for bactea-ia.
wexe 4..6,
~. rrae~-e~ter~iasr~a; 5.8, llf: lute~cs, and 6.8 ~'. c~la. The detection
limits will be governed
by the DMS through the balance between resolution and ion yield, which are
inversely related and controlled primarily by separation field. That is,
increased
resolution attained through increased separation field results in this
generation of
DMS with increased ion losses. In limited studies to confirm these
expectations,
separation fields were decreased and 6000 bacteria were detected for
E°. a~li using a
single biomarker.
EXAMPLE S
COMPARISON OF IMS-TOF TO DMS: RESOLUTION OF META- AND PARA XYLENES
To illustrate the advantages of the method and apparatus of the invention,
compounds that are extremely difficult to resolve in tune-of flight ion
mobility
spectrometry (TOF-I1VIS) are shown to be easily resolved in DMS practices of
the
invention herein. TOF-IMS is a highly sensitive, quantitative method for
organic
compound detection. It has been used for detection of chemical warfare agents,
illicit drugs and explosives, and unlike mass spectrometry, it operates like
the
present invention at atmospheric pxessure, eliminating the need for vacuum
tight
seals and power consuming vacuum pumps.
However, the TOF-IMS operates with low strength electric fields where the
mobility of an ion is essentially constant with electric field strength, while
the
present invention operates in periodic high fields and filters based on the
non-linear
mobility dependence of ions on the high strength fields. Thus the invention
can
provide more and different structural information about ion species that
further
enables accurate species detection and identification. Further comparison
v~ith'T~F-
IMS is instructive.
It will be understood that mixed xylenes are the second-most-important
aromatic product for chemical manufacturing around the world, ranking behind
benzene and ahead of toluene. Of the three isomers (ortho-, mete- and pare-
xylene)
p-xylene is the most widely used isomer in the manufacture of polymeric
materials.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-63-
Separation of these isomers is generally challenging with most analytical
instruments. Since these isomers have the same molecular weight they cannot be
resolved in a mass spectrometer.
In conventional TGF-II~S these compounds have virtually overlapping
peaks, as shown in FIG. 29. while these compounds can be resolved in a gas
chromatograph (GC), this typically takes more than 20 minutes. Meanwhile
devices
in practice of the invention enable excellent resolution of the pare and mete
xylenes
in under one second. FIG. 30 shows these compounds clearly resolved in
practice of
the invention, notwithstanding such surprisingly rapid performance.
EXAMPLE 9
HIGH SENSITIVITY OF DMS IN DETECTION OF METHYL SALICILATE
FIGS. 31 and 32 show the xesponse and concentration dependence in DMS
practice of the invention for methyl salycilate, a chemical warfare agent
simulant.
FIG. 31 shows positive ion spectra for different concentrations of methyl
salycilate.
FIG. 32 shows concentration dependence of the system to methyl salycilate for
both
positive and negative ions. Samples with concentrations of methyl salycilate
down
to about 45 parts-per-trillion are readily detectable in this device. The
methyl
salycilate compound produces both positively and negatively charged ions which
exhibit similar concentration dependences. The apparatus of the invention is
able to
simultaneously detect both ion responses within the same analytical run.
Producing
simultaneous positive and negative ion species information improves compound
identification at reduced detection times.
EXAMPLE 10
SMOKE ANALYSIS FROM THE COMBUSTION OF COTTON, PAPER, GRASS, TOBACCO
AND GASOLINE SAMPLES USING GC-DMS
lVlcztef°iczls eyed Il~letlt~c~s
Smol~e from combustion of cotton, paper, grass, tobacco and gasoline (in an
internal combustion engine) were sampled by SPME and the samples were screened
using a GC-DMS. As a control, a measure of the chemical vapor composition of
several materials using GC-MS was performed, as well as discrimination of
vapor
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-64-
profiles between the tested anaiytes (including several cellulose materials)
by GC-
DMS, and then identification of chemical markers specific to the burning of a
particular material.
In one demonstration of the invention, model 5880 gas chromatograph
(Hewlett-Packard Co., Avondale PA) was equipped with a HP splitless injector9
25
m SP 2300 capillary column (Supelco, Bellefonte, PA), a flame ionization
detector,
and a DMS detector. A model 5880 gas chromatograph (GC) (Hewlett-Packard Co.,
Avondale PA) was equipped with a HP splitless injector, 25 m SP 2300 capillary
column (Supelco, Bellefonte, PA), a flame ionization detector.
The Garner gas was nitrogen (99.99°/~) scrubbed over a molecular
sieve bed
and pressure on the splitless injector was 10 psig with a split ratio was
50:1. Other
experimental parameters for the GC included: iutial temperature, 30°C;
initial time,
5 min; program rate 15°C min-1; final temperature, 200°C; final
time, 1 min.
The DMS detector was equipped with ~0.6-1 mCi of 63Ni. The drift gas
was air at 1 to 21 min-1 from a model 737 Addco Pure Air generator (Miami,
FL).
The drift gas was further purified over a 5A molecular sieve bed (10 cm
diameter X
0.6 m long) and passed through heated stainless steel tubing to warm the drift
tube to
70°C.
The analytical colmnn was attached to the DMS drift tube through a 30 cm
length of aluminum-clad column and column effluent was carried by drift gas
through the ion source region for sample ionization. The drift gas also
carried
product ions through the gap (0.5 mm) between two flat separating electrodes
(5 x
mm). Two electric fields were applied to the drift tube: a non symmetric
wavefonn high frequency (1.3 MHz) with strong electric field (201cV cm-1 peals
to
25 peak amplitude) and a weak DC field (-360 V cm-1 to +80 V cm-1) of
compensation
voltage (-I8 to +4V). Signal was processed using a National Instruments board
(Model 6024E), digitized and stored. Excel 97 (Microsoft Corp) and ~rigin v
5.0
were used to display the results as spectra in topographic plots and graphs of
ion
intensity versus time.
The gas chromatograph-mass spectrometer was a model 5890 A gas
chromatograph and Model 5971A mass selective detector (Hewlett-Packard Co.,
Palo Alto, CA) and was equipped with a 25 m SP 2300 capillary column (Supelco,
Bellefonte, PA). The operating parameters of the gas chromatograph/mass
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-65-
spectrometer (GC/MS) were identical~to those for GC-DMS listed above.
Conditions for the mass spectrometer were: mass range, 45-550 amu; threshold,
500;
scan rate, 200 amu s-1; and electron multiplier voltage, 2100-2500 according
to the
automated calibrati~n routine.
Solid phase micro-extraction (Shl~~IE) fibers and injector were obtain ed from
Supelco (l3ellefonte, PA). h mixture of hydrocarbons (hexane to hexadecane)
was
prepared in methlyene chloride solvent at 100 ng/ul per alkane. The alkanes
were
obtained from various manufacturers and were used as a standard for
calibration of
chromatographic retention. Materials were all obtained locally and included
paper
as shredded newspaper; cotton; tobacco as cigarettes; and grass as dried
l3ermuda
grass.
Procedure
In one demonstration, a wad of ~9 cm3 of loosely held material (cigarette
excepted) was placed in the end of a borosilicate glass tube (2.54 cm OD X 6
cm
long) which was held level and a flame from a butane lighter was used to
ignite the
sample. The apparatus was placed in a fume hood where flow of air created air
flow
through the tube and allowed a sustained but low level burn of the sample over
3-~
minutes. Hot vapor and particulate emissions from the sample were released in
a
plume from the sample and the SPME fiber was held in this plume simulating
field
sampling of ambient air.
The time of sampling was 4 s for cotton, 6 s for cigarettes, S s for paper and
10 s for grass. Samples of engine exhaust from a forklii~ truck were taken by
holding the SPME fiber in the exhaust stream approximately 0.5 meter from the
end
of the tailpipe. The samples were freshly analyzed by GC-MS or GC-DMS. In an
injection, the SPME was placed in the injection port under splitless mode and
held
for 30 seconds until the inlet was switched to split mode. The SPME fibers
were
conditioned between runs for 10 minutes at 220°C in nitrogen.
Repeatability was
obtained by four replicate measurements of cotton burns with 4.s sampling of
the
smoke plumes. The all~ane standard was used to calibrate retention on the GC-
MS
and the GC-DMS.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-66-
Detection of V~OC f~ofn Combustion of Fuel Sources by GC MS
A preliminary requirement in this study was to determine if smoke samples
taken by SPME methods and analyzed by capillary GC-MS would provide
chromatographic profiles for V~Cs sufficiently distinct to be attributed to
specific
fuel sources. F~esults from GC-MS analysis of SPME samples from four of the
five
combustion sources (cotton, paper, grass and engine exhaust) axe shown in FIG.
33
as total ion chromatograms. These V~Cs spamled the range of carbon numbers
from 10 t~ 18 as shown in retention times for the alkane standard under
identical
conditions. Time (in minutes) for the all~anes (not shown) were: decane, 4.27;
undecane, 5.54; dodecane, 7.00; tridecane, 8.40; tetradecane, 9.81;
pentadecane,
11.17; hexadecane, 12.47, and octradecane, 15.79 (alkanes with carbon numbers
below 10 were lost in the solvent delay). The traces in FIG. 33 spanning 2 to
20
minutes retention illustrate that all samples exhibited a complex mixture of
VOCs
from adsorbed aerosols (desorbed in the injection port) and molecular weights
for
these compounds can be estimated as 150 to 250 amu. Additional chemical
information for compounds with molar masses belov~ 150 amu was not sought as
such compounds were regarded as too volatile to be collected by SPME sampling
methods. Also, no particular effort was made to measure constituents above 250
amu. The emphasis in these measurements was a comparison of emission
composition available by SPME sampling without requirements for cryogenic
options or ultra high temperatures. Thus, these findings are not a
comprehensive
chemical characterization of vapors and the measurements were made in
anticipation
of practical, field-portable instruments which would be engineered for
operation
under simple conditions according to embodiments of the invention.
In the range of molecular weights screened in FIG. 33, clear differences
existed in the qualitative and quantitative distributions of peaks in the
chromatographic profiles. While the profiles of total ion chromatograms appear
distinctive in FIG. 33, peaks in cotton, paper and grass were shared in common
by
these cellulose based materials though differences could be found in relative
abundances. An inspection of the chromatograms showed that there were 10
constituents in cotton that were distinctive over all other constituents in
other
samples and these distinctive components are shown in Table 1. In summary,
these
findings demonstrated that the chemical composition of emissions from bunting
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
_ 67 -
materials of interest exhibited measurable chemical differences through
analysis by
high resolution GC-MS. Thus, there is a chemical basis for an advanced smoke
detector to discriminate between source materials of fires. Naturally,
exhaustive
studies on the composition of smoke and reproducibility of sampling and
analysis
would be needed to fuxther refine these observations. however, precision
(discussed
below) demonstrated that the differences were not random, encouraging
f~.u'ther
study.
Ion mobility spectrometers are equipped with an atmospheric pressure
chemical ionization (A,FCI) source and an intermediate step was to determine
if the
pre-separation could be eliminated from the method. Direct sampling of
emissions
using an APCI mass spectrometer was made to determine the APCI response to
effluent constituents and to measure the resolution possible with a mass
spectrometer alone.
The mass spectra from direct sampling of vapors with a corona discharge ion
source for grass, cotton and cigarettes are shown for positive ions in FIG. 34
and
Table 2. In the background air, the reactant ions were ions with mlz 19, 35,
55, 73
aanu corresponding to ions of H30+(H2~)n with n=0,1,2,3, respectively.
Distinctive among these spectra for direct sampling of combustion emissions is
that
for cigarette smoke where nicotine is evident at m/z 163 amu. Nicotine has a
large
proton affinity and has been knoml fox decades to yield protonated monomers
through reactions as shown in Equation 3:
M + H3p~(H~Q)n ___________> MH+(H2p)n -r HBO (3)
vapor reagent ions product
ions
Consequently, nicotine preferentially acquired charge from the reactant ions
over other sample vapors and became the dominant ion through competitive
charge
exchange. ~ther constituents are present (as shown in FIG. 349 cigarette
smoke) but
the nicotine protonated monomer towers above all other peaks in the .P~CI mass
spectrum.
In the other samples, the distribution of vapor concentrations and proton
affinities of the VOCs yielded complex mass spectra with masses between 60-200
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-68-
amu for grass and cotton. There were complex with ions common to each owing to
gas phase ionization reactions at ambient pressure. Though multivariate
methods
might be employed to categorize the sources responsible for the mass spectra
in FIG.
34 (cotton smoke and grass smoke), the peaks were separated by unit mass
generally
in the mass spectrometer (Table 2). The resolution of a mobility spectrometer
is
inadequate to provide satisfactory separation of such complex ion mixtures.
Therefore, pre-separation, such as by capillary GC, was regarded as essential
for
highly reliable separation and identification.
(8C-I~lVIS analysis eaf Tj~C fr~m c~mbustioya ~~''va~i~us mczte~ials
While samples may be drawn directly from the ambient environment into the
first separation stage S-A in embodiments of the invention, we now discuss GC-
DMS analysis of SPME-collected samples of combustion emissions, as shown in
FIGS. 35 and 36 as chromatograms of total intensity of product ions from the
mobility scans in comparable format to total ion chromatograms from GC-MS (see
FIG. 33). Plots are shown for cotton, paper grass, cigarette and gasoline
engine
smoke and the findings reflect the same level of complexity (i. e., number of
resolved
constituents) as seen above in the GC-MS plots. Here also, as expected, the
VOCs
spamzed the range of carbon numbers from 10 to 18 as shown in retention times
spanning 3 to 15 minutes. For the alkane standard (not shown) under identical
conditions times (in minutes) were: nonane, 3.54; decane, 4.52; undecane,
5.81;
dodecane, 7.24; tridecane, 8.77; tetradecane, 10.27; pentadecane, 11.78; and
hexadecane, 13.43. The run was ended before octadecane eluted; alkanes with
carbon numbers below nine appeared in a large unresolved peak from 2 to 3.5
minutes.
Differences were observed in relative abundances of constituents within the
chromatograms between FIG. 34 and those in FIGS. 35 and 36. This is associated
with differences in response factors between mass spectrometry and ion
mobility
spectrometry or differences between vacuum based ion formation and ionization
at
atmospheric pressure. In the later, response is roughly approximated by proton
affinities of molecules. Thus, what appear as minor constituents in emissions
from
cotton between 8 to 10 minutes (FIG. 33) appear as significant constituents at
the
same retention time in FIG. 35. On the one hand, cigarette smoke nicotine
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-69-
dominated the DMS chromatographic response (FIG. 36) as it did the MS trace
(FIG. 34). On the other hand, the response to small molecules from C10 to C14
by
DMS (FIG. 36) was clearly more pronounced than that from MS (F'IG. 34).
Reproducibility wa.s determined using the peak heights on the product ion
plots of FIGs. 35 and 36 and results are shown in Table 3 for several peaks
from
throughout the elution pr ogTam. The relative standard deviation of the
measurement
was ranged across a comparatively narrow gap from 17 to 30 % RSD. This
variati~n included all aspects of sample preparation, sampling, measurement,
and
automated data reduction. As a consequence, results from this method exhibited
actual differences that could not be attributed to variance or random error
even
though sampling was made with comparatively casual attention to the
limitations of
SPME.
The peaks unique to cotton were labeled in FIGs. 33 and 35 and summaries
of the mass spectral properties for these are shown in Table 1. The GC-DMS
results
could be compared directly to the findings from GC-MS and certain peaks were
found in the cotton to be unique or special to cotton combustion. These are
labeled
in FIG. 35 and mass spectral properties are listed in Table 2. The 2D plots
are
particularly good for quantitative measures but do not disclose the analytical
value
of orthogonal information available in the mobility scan. This can be seen in
the
topographic plots of FIGS. 37 and 38.
Topographic plots from GC-DMS analysis of all samples demonstrated that
information in the mobility scan provides distinctiveness for each sample. The
most
distinctive of the plots, shown in FIGs. 37 and 38, is that from the internal
combustion engine where incompletely combusted hydrocarbons appear in a narrow
band of compensation voltage from -2 to 2 V. This was consistent with mass
spectra and with the alkane standard with peaks of similar compensation
voltage.
Plots for the cellulose materials (cotton, paper, and grass) exhibited some
common
features with peaks from 5 to 14 minutes evenly distributed and compensation
voltages that spurned -10 to +5V. In general, the compensation voltage trended
from -l OV toward OV with increased retention time. These results were
promising
as an examination of the concept of GC-DMS as an advanced smoke detector. As
well, it is further noted that cotton exhibited unique or characteristic peaks
in the 3-
D plots as labeled in FIG. 37 (cotton) and FIG. 33 (cotton).
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-70-
All of these results had been obtained using comparatively slow GC
temperature ramps. The 15 minutes might be reduced substantially with a fast
GC.
An essential question is what information is available in the plots and to
what degree
can this information be compressed without 1~sing measurexxmnt resolution
beyond a
usable condition. In mass spectrometry, selected ion monitoring is used to
improve
detection limits and can add selectivity through ion ratio measurements. This
saane
concept can be applied to DMS to monitor certain ions. This approach to
measurement provided a reliable route to fast chromatography (FIGS. 39 and
4~0).
W FIGs. 39 and 40, plots of ion intensity at four compensation voltages
(4.06'V, -l.SV, -4.47V, and-10.35V) are shown top to bottom in each frame for
cotton, paper, grass and gasoline engine exhaust. With this approach which is
analogous to ion chromatograms in GC/MS measurements, certain regions of
compensation voltage were graphically extracted from the matrix of retention
time,
compensation voltage and ion intensity. A measure of the amount of chemical
information in the selected ion mobility plots from GC-DMS can be seen in
these
figures where differences in samples are further accentuated. These patterns
are
distinct and show that selected ion plots allow a route to chemical
identification
where information can be compressed through fast chromatography. A further
embodiment includes forming ratios of two or more extracted ion chromatograms.
The findings in FIGS. 39 and 40 demonstrate that adequate resolution of
peals is available in these complex patterns to compress the chromatographic
time
scale. At present the scan time of 1 s may be too slow for below 3 minutes.
Since
ion residence in the drift tube is 1-2 ms, ion hopping could occur for a set
of ions in
40 ms, (4 ions for 10 ms each). Thus, high speed GC where the complete
separation
occurs in 60 seconds is preferred to allow high cycles (e.g., 1000) throughout
the
measurement. The time resolution of 1 part in 1500 would be comparable to the
current value of 1 in 960. Thus, should all other facets of separation be
scalable, the
drift tube with ion hoping enables high speed GC-DMS as a realistic smoke
detector
to distinguish sources of smoke.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-71-
Table 1. Retention times and ions in mass spectra from peaks that appear
distinctive to cotton emissions.
Peak Retention Prominent Ions in order of abundance (base
~To~~ Time peak abundance)
(min)
1 5.51 4~3 (190,000) 72, 55, 83, 98
2 7.13 42 (180,000), 41, 55, 86, 96
3 9.09 55 (24,000), 126, 41, 42, 4~3, 53, 69
4 9.55 42 (42,000), 41, 57, 56, 100
9.94 43 (19,500), 4~1, 55, 57, 709 69, 83
6 10.19 44 (390,000), 57, 43, 41, 128
7 11.78 69 (110,000), 57, 41, 42, 70, 43, 144
8 12.61 102 (54,000 , 132, 101, 789 77, 51, 50
'Refer to FIG. 33.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-72-
Table 2. Ion masses from direct sampling of smoke from cigarettes, grass, and
cotton with
analysis by atmospheric pressure ionization mass spectrometry
ll~Iass Cigarettes Grass Cott~n
Intensity Intensity Intensity
37 10,064,000 5,420,000
44 1,210,000
55 6,562,000 4,558,000
60 1,070,000 4,050,000 6,978,000
70 732,000
71 1,290,000
73 1,054,000
74 866,000
75 1,812,000
80 1,020,000 2,532,000
81 1,010,000
83 1,546,000 1,340,000
85 5,398,000 1,276,000
87 3,416,000
88 728,000
90 782,000
94 2,144,000 1,528,000
97 5,890,000 5,960,000
99 3,986,000 1,630,000
I 01 2, 104,0(30
102 2,020,000
103 3,338,000
104 1,102,000
106 1,134,000
108 930,000
109 1,548,000
111 2,874,000 6,284,000
113 2,010,000 2,042,000
115 2,860,000
116 1,558,000
1 I7 2,960,000
118 996,000
123 2,244,000
125 3,160,000
127 1,694,000 1,912,000
129 1,068,000
130 990,000
137 1,430,000
139 1,424,000
142 866,000
164 ~ 7,034,000
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-73-
Table 3. Reproducibility of combustion experiments including sampling, GC-MS
determination and data reduction. Measurements were made using 4 s burns of
cotton with
four complete replicate experiments.
Retention Time Area Standard % Relative Standard
(min) I~e~riation De~ation (l~l~SI~~
2.9 20696587 51884.74 25.07
26439681 5973165 22.59
8 19698838 6011899 30,52
11.8 12360466 3020822 24.44
12.7 10095391 1749393 17.33
14.02 3873334 1167082 30.13
EXAMPLE 11
DETECTION OF CHEMICAL WARFARE AGENT SMTLANTS
With the continuing threat from chemical and biological weapons, the need
for more effective and reliable detectors continues to be an issue for both
the
military and homeland security. Most, if not all, of today's deployed
detection
devices were developed to address the relatively narrow range of classic
warfare
agents (CWAs) of the cold war era. However, with the escalation of world
terrorism
there comes the need to deal with a broader range of threats that include a
substantial
list of toxics, including so-called toxic industrial chemicals and toxic
industrial
materials (a.k.a. TICs and TTMs). This places an even greater burden on
detector
technologies wluch must offer even higher selectivity without compromising
sensitivity. The requirement is for fast response times with significantly
lower false
positives. Most of the currently deployed detectors are based on IMS
technology,
developed to maturity over the last several decades and now struggling to
adapt to
these increasinglchanging requirements. Meanwhile, practices of the present
invention overcome these difficulties.
There are a variety of different interferences present in real world
conditions
such as: Aqueous Fire Fighting Foam (AFFF), diesel fuel, gasoline, pesticides,
paints and floor waxes that lead to a high rate of false positives in the
currently
deployed conventional IMS detectors. Frequent false alarms are ~ften
experienced in
the dusty, smol~e-ridden, enviroiaments. These lead to a loss of confidence in
detection equipment. These false positives can be caused in IMS equipment by
the
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-74-
fact that many ion species can have the same, or very similar, low field
mobility
coefficients. Practices of the present invention overcome these difficulties
as well.
In one practice of the invention, trace compounds were detected after
ioni~,ation with a 631~T1 radioactme source. To demonstrate the invention,
e~~periments
were perfonned with calibrated standards of the C~I-~ simulants: Dimethyl
methylphosphonate (DMMP), Diisopropylmethylphosphonate (DIMP) and Methyl
Salicylate (MS). Three independent vapor generator systems (Vici Metronics
Inc,
Model 190) were used to generate controlled air mixtures of the simulants at
different concentrations. Permeation sources were purchased fiom I~IN-TEI~
with
calibrated emission rates of 160 ng/min DMMP at T=HOC, 301 ng/min DIMP at
1000, and 5240 ng/min MS at 100C. Gas flow rates in all three systems were the
same 100 cc/min. The maximum sample concentrations that could be provided was
1.6 mglm~ for DMMP, 3.01. mglm3 for DIMP, and 52.4 mg/m3 for MS.
The DMS filter was operated in one aspect where it sampled effectively
100% of an incoming trace gas sample. In a 100% duty cycle aspect of the
invention, the compensation voltage is fixed such that a particular ion
species,
identified by its differential mobility, is permitted to reach the detector.
This is in
contrast to conventional iMS which typically uses a gate which is pulsed
"open" for
approximately 1 % of a measurement cycle resulting in only about 1 % of the
ions
being sampled. The DMS filter of this embodiment did not contain an ion-gate
as in
IMS devices, and is therefore more sensitive than gated approaches (where
large
portions of the sample signal axe discarded). In addition to the absence of a
gate, in
the present invention it is possible to improve sensitivity wherein the signal
can be
integrated over a relatively long period of time.
The DMS was also operated in a second aspect in order to produce a
spectroscopic output by scanning a range of compensation voltages. This
reduces the
"effective" duty cycle, but since the range of compensation voltages that is
scanned
can be selected by the operator, the "effective" duty cycle for any type of
ion species
is significantly higher than in conventional II~1S. The sensitivity of the
system is
higher than conventional IMS with the ability to detect compounds in the ppt
range.
Increased sensitivity is invaluable, especially in applications requiring
miosis
level detection. However, a detector which has high sensitivity without
selectivity
leads to an even higher rate of unwanted false positives. As previously
mentioned,
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-75-
enhanced selectivity in the DMS systems of the invention are provided by
changing
the electric field strength applied to the ionized molecules. In pxactice,
this translates
to changing the field strength of the asymmetric oscillating RF field (Vrf).
Changing
the I~ field results in a corresponding shift in the spectral peak position,
as
measured by the compensation voltage. Changing the RF field leads to tunable
resolution accessed by changing the I2F filtering amplitude and thus changing
the
operating point on the mobility vs. electric field curve. In the system the
various RF
field values can all generated under the automatic control of the
microprocess~r. The
tunable resolution makes it possible to separate monomers fiom dimers and
other
clusters and to use these cluster peaks to aid in the identification of
compounds, for
example, at FIG. 41 (top curves).
Tunable resolution also enables the RIP (reactant ion peak) to be displaced
away from the peaks of interest. The RIP is a background peak that frequently
interferes with the detection of targeted compounds in conventional IMS. This
property of the invention enables the detection of trace compounds in
backgrounds
that produce interfering signals, or in some situations where the RIP or other
compounds would otherwise interfere with successful detection. Tracking how
the
spectral peak position shifts with RF field provides a great deal of
information
unique to that ion species, peaks for other ion species will shift very
differently.
These are species-specific signatures which can be used to identify detected
species
in practice of the invention.
As earlier described, embodiments of the present invention can
simultaneously detect both positive and negative ion peaks, modes, which
further
helps to improve selectivity. The absence or the presence of peaks, and their
size,
and location, in the positive ion channel versus the negative channel provides
more
information on the specific compound identity. The ratios between intensities
of
positive ions and negative ions for a given sample also provides additional
information which enhances confidence of the detection.
FIG. ~2A and FIG. 42~ illustrate this for the nerve agent GA. These spectral
plots were measured at a Vrf of 1,4S2v which corresponds to a field strength
of
2~,64.Ov/cm. The characteristic spectrum for the positive ions is very
different from
the negative ion spectrum.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-76-
The combination of the positive and negative ion channel information,
together with the information provided by monitoring the spectral peak shifts
as a
function of the applied RF held, results in a powerful tool for chemical
identification
in practice of the present in~~ention.
In soame cases it is desired to achieve narrower peaks and better peak
resolution, such as for discriminating between peaks for similar or
interfering
analytes in a sample. An additional embodiment of the present invention
addresses
this concern by operating the system at slightly reduced pressures relative to
atmosphere. Under these reduced pressure conditions, down to 0.5 atmospheres,
the
resolution according to the invention is significantly increased.
The effect of reduced pressure is illustrated in FIG. 43 for three CWA
simulants, DMMP, DIlVIP, and MS. The top spectra show the results obtained at
atmospheric pressure, RF=1000V, while the next spectra was obtained at 0.65
atm,
RF=800V, and the hottom spectra was obtained at 0.5 atm, RF=650V. It will be
appreciated that the top scan discriminates between the three simulants with
some
spectral overlapping but which may be adeduate in some cases. However the next
lower scan has better resolution (narrower peaks) and the lowest scan has even
better
resolution. Thus it is possible to discriminate between such analytes in a
sample by
reducing operating pressure, according to aspects of the invention.
A further advantage of reducing pressure in the system is that the amplitude
of RF voltages required to filter the ions can be reduced, this results in a
lower
power requirements which is especially important for field-portable systems of
the
invention.
In one embodiment of the invention, direct sampling of volatile chemical
agents provides adequate detection results, as was the case with FIG. 41,
especially
in the upper scans. However it is also possible to use a first stage of
separation, such
as previously discussed which results in providing a less complex ionized
sample to
the DMS, resulting in a less complex sample being ionized and filtered in the
DMS
filter.
In one practice of the invention, a membrane was used at the input of the
DMS filter system prior to ionization, such as at separation stage S-A in FIG.
2A.
The plots of FIG. 43 were obtained with such a membrane front end. Selection
of
membrane is guided by the need to selectively pass CWA agent molecules while
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
acting as a barrier to moisture and heavier molecules (whether dirt, dust,
hydrocarbon exhausts or the like). Thus a better and more controlled
analytical
envixonment can be provided within a detection system of the invention with a
less
complex sample being aoni~ed and altered vrithin the DMS filter. The Flots of
FIG.
43 were taken using a membrane front end as separator that selectively
introduced
the detected CWA agents as indicated while acting as a barrier to moisture and
unwanted heavier molecules. Various membrane materials are knovm in the art,
including partially porous materials. These materials may include Teflon,
latex,
pdms, dimethyl silicone, or the like, as may be used in membrane practices of
the
invention.
One of the critical aspects of a CWA detector is how well it can reject
interferants to prevent false alarms. One particular interferant, Aqueous Fire
Fighting Foam (AFFF}, has proved extremely challenging for conventional IMS to
resolve from CWAs, or CWA simulants. The AFFF peak tends to overlap with that
of the agent peals in IMS. FIG. 44 demonstrates a practice of the invention
for a
series of warfare agent simulants selectively mixed with 1 % headspace of
AFFF. As
can be seen, good DMS peak resolution can be achieved in practice of the
invention.
There are eight spectral plots in FIG. 44. The top plot shows the RIP for a
DMS
system with background air but no sample present with the sensor at
atmospheric
pressure. The next plot shows AFFF interferant added. This results only in a
slight
shift to the left (more negative compensation voltage) of the RTP peak. The
CWA
simulant DMMP is then introduced alone into the spectrometer and the typical
monomer and dirnlner peaks appear together with a corresponding reduction in
the
RIP peals intensity. When 1% AFFF is added, the DMMP peaks are not effected
and
only a slight leftward shift of the RIP is observed. The same experiment was
repeated with DIMP and the effect of AFFF was negligible. Introducing MS and
monitoring the negative ion peaks gave the similar data illustrating the Lack
of
interference with AFFF. The conclusion is that I °f~ AFFF has virtually
no effect on
the DMS practices of the invention for CWA simulant spectra. Similar results
were
obtained with live agents as well. This is an important breakthrough for CWA
monitoring.
CA 02518703 2005-09-08
WO 2004/081527 PCT/US2004/007213
-7S-
The present invention enables method and apparatus for high field
asymmetric waveform ion mobility spectrometry, which can be favorably
augmezlted with other collection and separation techniques, and packaged in a
compact system. ~ther than planar configurations axe possible. I~I~S
co~nfigur~.tions
which may be practiced according to the invention may include method and
apparatus uSlllg co-axial, cylindrical, flat, planar, radial, curved and other
I~I~IS
elects~de configurations. Embodiments of the invention may even be practiced
augmented v~ith prior art Il~IS and I~IVIS filtering.
The high sensitivity, rugged design and ease of use and setup of
embodiments of the invention are advantageous for many applications that
involve
chemical detection. A simplified hand-held device of the invention is
dedicated to
detection of a limited set of data and yet reliably detects and identifies ion
species of
interest. This practice may be augmented by dual mode detections. The result
is
added reliability in chemical detection in a simplified device.
It will now be appreciated that in practice of the invention we control the
filter field,
its electrical properties and its environment, its an ion-mobility based
system, to
amplify differences in ion mobility behavior for species separation. Species
are then
detected and identified based on this function. We can further optimize the
process
by controlling ionization (such as by selection of sources of lower or higher
levels of
ionization energy).
It should be appreciated that numerous changes may be made to the
disclosed embodiments without departing from the scope of the present
invention.
While the foregoing examples refer to specific embodiments, this is intended
to be
by way of example and illustration only, and not by way of limitation. It
should be
appreciated by a person slcilled in the art that other chemicals and molecules
may be
similarly ionized and detected.
Therefore, while this invention has been particularly shown and described
with references to the above embodiments, it will be understood by those
skilled in
the art that various changes in form and details may be made therein v~ithout
departing from the scope of the invention encompassed by the appended claims.
