U.S. patent application number 12/036045 was filed with the patent office on 2009-03-26 for compact cross-dispersed spectrometer for extended spectral range.
This patent application is currently assigned to THERMO NITON ANALYZERS LLC. Invention is credited to John E. Goulter, Mark Hamilton, Pratheev Sreetharan.
Application Number | 20090079980 12/036045 |
Document ID | / |
Family ID | 39580291 |
Filed Date | 2009-03-26 |
United States Patent
Application |
20090079980 |
Kind Code |
A1 |
Goulter; John E. ; et
al. |
March 26, 2009 |
Compact Cross-Dispersed Spectrometer for Extended Spectral
Range
Abstract
A spectrometer includes a structural member made of a
light-weight material having a small coefficient of thermal
expansion (CTE). The spectrometer is dimensionally stable over a
range of expected ambient temperatures, without controlling the
temperature of the spectrometer.
Inventors: |
Goulter; John E.;
(Northridge, CA) ; Hamilton; Mark; (Upton, MA)
; Sreetharan; Pratheev; (Medford, MA) |
Correspondence
Address: |
THERMO FINNIGAN LLC
355 RIVER OAKS PARKWAY
SAN JOSE
CA
95134
US
|
Assignee: |
THERMO NITON ANALYZERS LLC
Billerica
MA
|
Family ID: |
39580291 |
Appl. No.: |
12/036045 |
Filed: |
February 22, 2008 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
60891408 |
Feb 23, 2007 |
|
|
|
Current U.S.
Class: |
356/326 ;
359/819 |
Current CPC
Class: |
G01J 3/0286 20130101;
G01J 3/0272 20130101; G01N 21/67 20130101; G01J 3/0297 20130101;
G01J 3/0256 20130101; G01J 3/0202 20130101; G01J 3/02 20130101;
G01J 3/2803 20130101; G01N 21/718 20130101; G01J 3/0264 20130101;
G01J 3/36 20130101; G01N 21/33 20130101; G01J 3/0283 20130101; G01J
3/0291 20130101; G01J 3/12 20130101; G01N 2201/0221 20130101; G01J
3/443 20130101 |
Class at
Publication: |
356/326 ;
359/819 |
International
Class: |
G01J 3/28 20060101
G01J003/28; G02B 7/02 20060101 G02B007/02 |
Claims
1. A method for mounting an optical element, comprising: providing
an optical mount comprising a carbon-filled polymer; and attaching
the optical element to the optical mount.
2. A method in accordance with claim 1, wherein the carbon-filled
polymer comprises graphite-filled polyphenylene sulfide.
3. A method in accordance with claim 1, wherein the carbon-filled
polymer comprises polyphenylene sulfide filled with at least about
40% carbon.
4. A method in accordance with claim 1, wherein the optical element
comprises a lens.
5. A method in accordance with claim 1, wherein the optical element
comprises an optical sorter.
6. A method in accordance with claim 1, wherein the optical element
comprises a diffraction grating.
7. A method in accordance with claim 1, wherein the optical element
comprises a prism.
8. A method in accordance with claim 1, wherein the optical element
comprises a mirror.
9. A method in accordance with claim 1, wherein the optical element
comprises a mask.
10. A spectrometer, comprising: a carbon-filled polymer structural
member; a light dispersion element mounted to the structural
member; and a sensor mounted to the structural member and oriented
to receive dispersed light from the light dispersion element.
11. A spectrometer in accordance with claim 10, wherein the
carbon-filled polymer comprises polyphenylene sulfide filled with
at least about 40% carbon.
12. A spectrometer in accordance with claim 10, further comprising:
an input; and an order sorter disposed between the input and the
light dispersion element.
13. A spectrometer in accordance with claim 12, further comprising
a structure defining an aperture disposed between the order sorter
and the light dispersion element.
Description
[0001] The present application claims priority from U.S.
Provisional Patent Application No. 60/891,408, titled "Hand-Held,
Self-Contained Optical Emission Spectroscopy (OES) Analyzer," filed
Feb. 23, 2007, which is incorporated in its entirety by reference
herein.
[0002] The contents of commonly-assigned U.S. patent application
Ser. No. 12/035,477, by Denis Baiko, et al., titled "Fast and
Precise Time-Resolved Spectroscopy with Linear Sensor Array," filed
Feb. 22, 2008, is incorporated in its entirety by reference
herein.
[0003] The contents of commonly-assigned U.S. patent application
Ser. No. ______ (to be supplied), by John E. Goulter, et al.,
titled "Hand-Held, Self-Contained Optical Emission Spectroscopy
(OES) Analyzer," filed Feb. 22, 2008, is incorporated in its
entirety by reference herein.
TECHNICAL FIELD
[0004] The present invention relates to spectrometers and, more
particularly, to compact spectrometers that remain dimensionally
stable and accurate over a temperature range without heating or
cooling.
BACKGROUND ART
[0005] Analyzing chemical composition of samples is important in
many contexts, including identifying and segregating metal types
(particularly various types of iron and steel) in outdoor metal
recycling facilities, quality control testing in factories and
forensic work. Several analytical methods are available.
[0006] Optical emission spectroscopy (OES) is a mature, robust
technology for the elemental analysis of materials. In OES, a small
quantity of sample material is vaporized and excited above atomic
ground state. Emissions characteristic of elements in the vaporized
sample are captured by a light guide, which sends the light to a
spectrometer, which produces and analyzes a spectrum from the
light, so as to yield the elemental composition.
[0007] For metal samples, the prevalent techniques for generating
an emission spectra use either an electric arc or a spark, or both,
to vaporize a small quantity of the sample to be analyzed.
Alternatively, laser-induced breakdown spectroscopy (LIBS) or glow
discharge (GD) may be used to vaporize and excite an emission
sample. A survey of OES analytical techniques may be found in
Slickers, Automatic Atomic-Emission Spectroscopy, Second Edition
(1993), which is incorporated by reference as if fully set forth
herein.
[0008] In order to be confident that the composition deduced from a
measurement, which typically tests a miniscule portion of the
sample, is representative of the composition of the entire sample,
minimizing effects from, for example, inclusions, matrixes and
surface contaminants, it is standard practice to average the
spectra from as many as several thousand arcs/sparks that have
struck an area as large as 100 square mm in a few seconds of a
measurement.
[0009] Some OES analyzers are large, non-portable units intended
for use in laboratories. Other OES analyzers are "portable," in
that they can be moved about. However, prior art "portable" OES
analyzers that can identify carbon or other common constituents in
iron or steel require two separate components interconnected by a
fiber optic/electric cable. For example, an analyzer available from
Spectro A. I., Inc. under the trade name Spectroport includes a
hand-held probe connected via a 10-foot cable to a suitcase-sized,
33-pound analysis unit. The Spectro iSort analyzer, also from
Spectro A. I., Inc., includes a hand-held probe connected by a
cable to an analysis unit housed in a 10-pound backpack.
[0010] To cover a spectral range required to detect carbon,
phosphorous, sulfur and other elements necessary to identify common
materials, such as cast iron and various alloys, these prior art
analyzers include fixed-wavelength detectors in the hand-held
probes for carbon, phosphorous, sulfur and iron, as well as a
spectrometer in the analysis unit for other elements. This awkward,
two-part structure makes these analyzers difficult to use and move
about.
[0011] An two-part analyzer available from Metorex, Ewing, N.J.,
under the trade name ARC-MET 8000 MobileLab, includes a hand-held
"probe" connected by a ten-foot cable to a roll-around "main unit."
The probe contains a spectrometer with an advertised spectral range
of 175-370 nm; however, the roll-around main unit is required to
provide power and cooling to the probe and to analyzes the output
from the spectrometer. At least some users would prefer to use a
hand-held OES analyzer that is fully self-contained.
[0012] The Spectrosort analyzer, also from Spectro A. I., Inc., is
a one-piece, battery-powered, hand-held analyzer. However, spectral
limitations of the spectrometer in this analyzer make it incapable
of detecting carbon, phosphorous and sulfur, thus severely limiting
the utility of this analyzer.
[0013] Users of self-contained, hand-held OES analyzers would
prefer analyzers that are capable of detecting carbon and other key
elements, so the analyzers can identify a wide range of common
materials. However, various roadblocks have thus far prevented
construction of such a full-range, self-contained, hand-held
analyzer. Among these roadblocks is an inability to construct a
spectrometer that exhibits the wavelength range and temperature
stability needed for the above-described analysis under typical
environmental conditions, in a size and weight appropriate for a
hand-held analyzer,
SUMMARY OF THE INVENTION
[0014] An embodiment of the present invention provides an analyzer
for analyzing composition of a portion of a sample. The analyzer
includes a hand-held, self-contained, test instrument. The test
instrument includes an exciter for exciting the portion of the
sample, the excitation producing an optical signal and a first
dispersive element disposed within the hand-held instrument for
receiving the optical signal and creating an intermediate optical
signal dispersed in a first plane. A second dispersive element
disposed within the hand-held instrument disperses the intermediate
optical signal so as to place a first resolved optical order on a
corresponding first plurality of detector elements and a second
resolved optical order on a corresponding second plurality of
detector elements. A processor is coupled to receive signals from
the first and second pluralities of detector elements and is
programmed to process the signals. A battery powers the exciter and
the processor.
[0015] At least one of the optical orders placed on the
corresponding plurality of detector elements may extend to
wavelengths shorter than about 193 nm, or shorter than about 178
nm, or at least as short as about 170 nm.
[0016] Each plurality of detector elements may be configured so as
to receive a continuous spectral range of the resolved optical
order placed on the plurality of detector elements. The spectral
range placed on the first and second pluralities of detector
elements may extend at least from about 178 nm to about 400 nm.
[0017] The instrument may include a structure defining an aperture,
through which the intermediate optical signal passes. The optical
signal may be focused on the structure.
[0018] The exciter may include an electrode for sustaining an
electrical potential with respect to the portion of the sample and
a voltage supply for establishing the electrical potential on the
electrode with respect to the portion of the sample. The exciter
may include a laser.
[0019] The first dispersive element may be a cross-dispersing
prism. The second dispersive element may be a diffraction grating,
which may be a holographic diffraction grating blazed to provide
comparable efficiencies in the first and second resolved optical
orders.
[0020] The first plurality of detector elements may be not
co-planar with the second plurality of detector elements. The test
instrument may further include a mirror in an optical path of one
of the first and second resolved optical orders, between the second
dispersive element and the corresponding plurality of detector
elements.
[0021] The first and second dispersive elements and the first and
second pluralities of detector elements may be rigidly coupled to a
carbon-filled polymer structural member, which may include
polyphylene sulfide filled with graphite, such as with at least
about 40% graphite.
[0022] The processor may be programmed for automatic wavelength
calibration, based on observed spectral features.
[0023] The second dispersive element may provide a resolving power
of at least about 5,000 or at least about 10,000.
[0024] The instrument may further include a display screen coupled
to the processor. The display screen may be a hinged display
screen.
[0025] An embodiment of the present invention provides an analyzer
for analyzing composition of a portion of a sample. The analyzer
may include a hand-held, self-contained, test instrument that
includes an exciter for exciting the portion of the sample. The
excitation produces an optical signal. The instrument also includes
a spectrometer having a spectral range extending at least from
about 178 nm to about 400 nm disposed in the analyzer to receive
the optical signal and operative to disperse the optical signal and
produce an output signal from the dispersed optical signal. The
instrument also includes a processor coupled to the spectrometer
and programmed to process the output signal and a battery powering
the exciter, the spectrometer and the processor.
[0026] The spectrometer may include a pixilated sensor, and the
spectrometer may have a resolution of at least about 0.02 nm per
pixel at about 190 nm.
[0027] The spectrometer may include a holographic diffraction
grating having comparable efficiency in at least two different
orders and sensors arranged to receive two orders of the dispersed
optical signal from the grating. The spectrometer may be
cross-dispersed.
[0028] The spectrometer may include a structural member that
includes a carbon-filled polymer, to which optical elements of the
spectrometer are mounted.
[0029] The processor may be programmed to automatically wavelength
calibrate the spectrometer, based on observed spectral
features.
[0030] Another embodiment of the present invention provides a
method for analyzing composition of a portion of a sample. The
method includes exciting the portion of the sample, thereby
producing an optical signal and generating a spectrum from the
optical signal. A first predetermined spectral feature is matched
with at least a portion of the spectrum. A wavelength is associated
with a pixel, based on a location of the first predetermined
spectral feature, relative to the pixel. The spectrum is analyzed
to determine at least one constituent of the portion of the
sample.
[0031] Wavelengths may be associated with other pixels, based on an
expected linear spectral dispersion over a set of pixels.
[0032] A second predetermined spectral feature may be matched with
at least a portion of the spectrum and wavelengths may be
associated with other pixels, based on a location of the second
predetermined spectral feature, relative to the location of the
first predetermined spectral feature.
BRIEF DESCRIPTION OF THE DRAWINGS
[0033] The invention will be more fully understood by referring to
the following Detailed Description of Specific Embodiments in
conjunction with the Drawings, of which:
[0034] FIG. 1 is a perspective view of a hand-held, self-contained,
battery-powered OES test instrument, according to an embodiment of
the present invention;
[0035] FIG. 2 is a cut-away view of the test instrument of FIG.
1;
[0036] FIG. 3 is a perspective view of a spectrometer of the
instrument of FIG. 1, according to an embodiment of the present
invention;
[0037] FIG. 4 is a close-up perspective view of a portion of the
spectrometer of FIG. 3;
[0038] FIG. 5 is shows a snout of the test instrument of FIG. 1 in
contact with a sample surface;
[0039] FIG. 6 is a graph showing representative voltage and current
curves plotted against time for a single spark/arc generated by the
instrument of FIG. 1, according to one embodiment of the present
invention;
[0040] FIG. 7 contains two graphs showing representative voltage
and current curves plotted against time for sparks/arcs generated
by the instrument of FIG. 1, according to one embodiment of the
present invention;
[0041] FIG. 8 is a close-up schematic view of an analytical gap
produced by the test instrument of FIG. 1;
[0042] FIG. 9 is a cutaway perspective diagram of the spectrometer
of FIGS. 3 and 4, according to one embodiment of the present
invention;
[0043] FIG. 10 contains two schematic representations of spectra
projected on sensor arrays of the instrument of FIG. 1, in relation
to selecting system parameters according to one embodiment of the
present invention;
[0044] FIG. 11 is a schematic diagram two rows of sensors,
according to one embodiment of the present invention;
[0045] FIG. 12 is a perspective view of the two rows of sensors of
FIG. 11 and corresponding mounting brackets therefore, according to
one embodiment of the present invention;
[0046] FIG. 13 is an exploded view of a diffraction grating
assembly, according to one embodiment of the present invention;
[0047] FIG. 14 is a perspective view of the diffraction grating
assembly of FIG. 13 mounted within a spectrometer housing,
according to one embodiment of the present invention;
[0048] FIG. 15 is a schematic diagram of an alignment setup for the
test instrument of FIG. 1, according to one embodiment of the
present invention;
[0049] FIG. 16 is a flow chart describing a process for analyzing
composition of a sample, according to one embodiment of the present
invention;
[0050] FIG. 17 is a block diagram of major components of the test
instrument of FIG. 1, according to one embodiment of the present
invention;
[0051] FIG. 18 is a perspective view of a hand-held, self-contained
test instrument with a tilt-up screen, according to one embodiment
of the present invention;
[0052] FIG. 19 is a schematic graph illustrating two spectral peaks
and corresponding signals produced by sensor pixels;
[0053] FIGS. 20 and 21 depict two embodiments of staggered pixel
structures for a sensor array, in accordance with embodiments of
the present invention;
[0054] FIG. 22 is a perspective schematic diagram of a sensor and a
spectrum and a shifted spectrum impinging on the sensor, according
to one embodiment of the present invention;
[0055] FIG. 23 is a graph illustrating a known spectrum; and
[0056] FIG. 24 is a flowchart that describes automatic wavelength
calibration, according to embodiments of the present invention.
DETAILED DESCRIPTION OF SPECIFIC EMBODIMENTS
[0057] In accordance with embodiments of the present invention,
methods and apparatus are disclosed for analyzing composition of a
sample with a hand-held, self-contained, battery-powered test
instrument. A spectrometer in the test instrument has a wavelength
range broad enough to enable the test instrument to detect and
determine relative quantities of carbon, phosphorous, sulfur,
manganese, silicon, iron and other elements necessary to identify
common alloys. The test instrument's design and construction
enables the test instrument to produce accurate results over a wide
ambient temperature range, without heating or cooling the
spectrometer, thereby conserving power and extending the amount of
time the test instrument may be operated before the battery needs
to be recharged.
[0058] The test instrument excites at least a portion of a sample,
thus producing an optical signal. As a result of optical emissions
at wavelengths characteristic of elements in the sample, the
optical signal contains information that identifies the elements in
the sample. The spectrometer wavelength-disperses the optical
signal onto a set of sensors, each of which receives a narrow range
of wavelengths of the optical signal. A processor is programmed to
receive and process signals from the sensors and to identify and
quantify the elements in the sample.
[0059] A hand-held, self-contained, battery-powered test instrument
should be small, light-weight and consume little electrical power.
Disclosed embodiments of the present invention enable construction
of spectroscopy-based analytical test instruments that exhibit
these properties. These embodiments are discussed in the context of
analytical techniques and test instruments that employ optical
emission spectroscopy (OES); however, the teachings of this
application are applicable to other types of analytical test
instruments that employ spectral analysis, including test
instruments that employ optical absorption spectroscopy.
Furthermore, although the disclosed embodiments are discussed in
the context of arc/spark excitation, other forms of excitation,
including laser-induced breakdown (LIB) and glow discharge (GD) may
be used.
General Structure of One Embodiment
[0060] FIG. 1 is a perspective view of a hand-held, self-contained,
battery-powered OES test instrument 100, according to one
embodiment of the present invention. The instrument 100 includes a
snout 102. In operation, an electrically-conductive flat portion
103 of the snout 102 is pressed against an electrically-conductive
sample surface (not shown). A spark from a counterelectrode 104 to
the sample excites a portion of the sample, thereby producing an
optical signal. The counterelectrode 104 is electrically insulated
from the electrically conductive flat portion 103 of the snout 102,
such as by an insulated disk (not visible). The optical signal
enters an upper port (not visible) and is reflected by one or more
mirrors (not visible) into a spectrometer 204 inside the instrument
100. A processor (not visible) is coupled to a set of detectors
(not visible) in the spectrometer. The processor is programmed to
process signals from the detectors. The processor analyzes at least
a portion of the spectrum produced by the spectrometer to identify
and quantify the elemental composition of the sample.
[0061] The processor displays results of the analysis on a
touchscreen 110. Optionally, the processor may transmit results of
the analysis to an external device, such as a computer or display,
via a wired or wireless connection (not shown). The touchscreen
110, a trigger 112 and operator interface buttons 114 enable a user
to interact with the processor. A detachable, rechargeable battery
116 powers the processor, touchscreen 110, spectrometer 204 and a
spark generator (not visible) that is coupled to the
counterelectrode 104.
[0062] FIG. 2 is a cut-away view of the test instrument 100 showing
the spark generator 200, a first mirror 202 and the spectrometer
204. The mirror 202 may be an aluminized front-surface mirror with
a magnesium fluoride coating and a fused silica substrate, although
other suitable mirrors may be used. The mirror 202 may be planar,
although a concave (including hyperbolic or parabolic) shape may
provide better image quality.
[0063] An electrically conductive insert 206 defines a bore 208, in
which the counterelectrode 104 (not shown in FIG. 2) is disposed.
The insert 206 also provides at least part of the electrically
conductive flat portion 103 of the snout 102. The spark generator
200 is electrically connected to the counterelectrode 104 and to
the electrically conductive flat portion 103 of the snout 102 to
complete on an electrical return circuit with the sample, when the
flat portion 103 of the snout 102 is brought into contact with the
sample. Much of the snout 102 may be metal or another
heat-conductive material to dissipate heat from the spark and from
the spark generator 200. A dashed line 210 schematically
illustrates a portion of a light path (largely hidden within the
snout 102) taken by the optical signal from the vicinity of the
counterelectrode 104 to an entrance slit of the spectrometer
204.
[0064] FIG. 3 is a perspective view of the spectrometer 204. The
spectrometer 204 housing includes a structure 300 that defines an
opening 302, into which the insert 206 (not shown in FIG. 3) in
inserted. FIG. 4 is a close-up perspective view of a portion of the
spectrometer 204 showing the structure 300 and the insert 206 in
more detail. The insert 206 defines a bore 400 through a wall of
the insert 206, as shown by dashed lines. The optical signal
produced by the sample passes through the bore 400 and is reflected
by a second mirror 402 in the structure 300. The second mirror 402
reflects the optical signal via a second bore 404 through the
structure 300 to the first mirror 202 (not shown).
[0065] A third bore 406 in the structure and a corresponding bore
(not visible) in the insert 206 provide a fluid communication path
through which a gas may be plumbed to the vicinity of the counter
electrode 104 to purge the vicinity of the counterelectrode 104 of
air, at least in part because the air may attenuate or block some
or all wavelengths of interest in an optical signal. A window (not
shown) mounted in the bore 404 provides a gas-tight seal to prevent
the gas from escaping from the bore 208 into the spectrometer 208.
The window is preferably made of beta alumina
(.beta.-Al.sub.2O.sub.3 or "synthetic sapphire") or another
material that is sufficiently transparent at the wavelengths of the
optical signal.
Environment
[0066] As shown in FIG. 3, a sharpened tip portion of the
counterelectrode 104 is disposed about 2-3 mm from the sample
surface 500, thereby creating an analytical gap. The
counterelectrode 104 may be about 1/16-1/4 inch in diameter. The
counterelectrode 104 is preferably made of thoriated tungsten,
although other suitable materials, such as carbon (graphite) or
silver may be used. The counterelectrode 104 should be made of a
material that produces a simple spectrum if excited, or at least a
spectrum that is easily distinguished from spectra produced by
likely materials in the sample 500.
[0067] An inert gas, such as argon, may be plumbed via the bore 406
to flood the analytical gap with the gas. Methods and apparatus for
providing a gas to a hand-held test instrument, including from a
gas tank coupled directly to, and possibly enclosed within a
portion of, the instrument, are disclosed in detail in Provisional
Patent Application No. 60/889,465, filed Feb. 12, 2007, titled
"Small Spot X-Ray Fluorescence (XRF) Analyzer," the contents of
which are incorporated by reference as if fully set forth herein. A
gas that is not chemically reactive with likely materials in the
sample 500, and that produces a relatively simple emission spectrum
when excited (or at least a spectrum that is easily distinguished
from spectra produced by likely materials in the sample 500),
should be selected.
Spark Generation
[0068] An electrical potential between the counterelectrode 104 and
the sample surface 500 breaks down the gas, enabling an electrical
current, in the form of a spark or an arc or both, to flow from the
counterelectrode 104 to the sample surface 500. The spark heats the
gas and vaporizes a small amount of the sample. The vaporized
sample material is excited by the hot gas and produces an optical
(although possibly invisible) discharge.
[0069] Positive unidirectional current should be provided to the
counterelectrode 104 to prevent eroding the counterelectrode 104.
The spark generator 200 includes a diode 508 (or an equivalent
circuit) to provide an appropriate unidirectional current to the
counterelectrode 104. The counterelectrode 104 may be cleaned of
debris buildup with a wire brush or by reversing the current and
producing sparks/arcs to a sacrificial cleaning sample.
[0070] In operation, a series of sparks/arcs may be generated in
rapid succession. Each spark may strike a slightly different
location on the sample surface 500, due to pitting of the sample
surface 500 by the sparks, imperfections and inclusions in the
sample surface 500, etc. In general, a high spark repetition rate
causes the sparks to strike the sample surface over a smaller area
than a low spark repetition rate causes. Thus, it is possible to
control the sample area by controlling the spark repetition rate.
At about 50 to 400 sparks per second, the sparks strike an area
about 3 mm in diameter, whereas at about 1,000 to 2,000 sparks per
second, the discharge area is about 1 mm in diameter. It may be
desirable to avoid small discharge areas, such as 1 mm in diameter,
at least in part because most metals are not sufficiently
homogeneous to yield accurate results when only such a small area
is tested. Sampling such a small area may produce a result that is
biased by the composition of the small area.
[0071] Voltage and current (versus time) profiles (waveforms) of a
signal provided to the counterelectrode 104 to produce the
spark/arc should be controlled to optimize initiating the spark,
vaporizing the sample and heating the gas, while limiting power
consumption. FIG. 6 is a graph showing representative voltage and
current curves plotted against time for a single spark/arc. As
shown in the graph, a short-duration, high-voltage peak 600
initiates a spark to breakdown the gas in an analytical gap between
the counterelectrode 104 and the sample surface 500. The spark
erodes a portion of the sample into the analytical gas. Thereafter,
the voltage is reduced. The spark is a low-current spark, as
indicated in a portion 602 of the current graph. However,
thereafter the current is increased and reaches a peak 604 while
the voltage is moderately high to sustain an arc to excite the
eroded sample material in the analytical gap. The excited material
emits an optical signal characteristic of the elemental composition
of the excited material. Thereafter, the current and voltage are
reduced. Due to the varying amounts of power introduced into the
analytical gap over the course of the spark/arc, the temperature in
the analytical gap varies over the duration of the spark/arc.
[0072] For analyzing hard metals, such as iron and nickel, a
voltage profile as shown in the top graph (A) of FIG. 7 may be
used. The voltage profile shows a high-energy pre-spark 700,
followed by a high-current, but lower voltage, arc 702. For
analyzing soft metals, such as aluminum, magnesium or copper, a
two-phase voltage profile as shown in the bottom graph (B) of FIG.
7 may be used. The voltage profile shows a spark portion 704, a
delay 706 and a separate arc portion 708. The spark portion 704 may
be used to determine the primary alloy in a sample, and the arc
portion 708 may be used to determine trace elements in the sample.
In general, the voltage of the signal in the lower graph (B) is
less than the voltage of the signal in the upper graph (A).
[0073] In one embodiment, the initial breakdown voltage is about
6,000-10,000 volts, as required to break down the argon or other
gas in the analytical gap. In one embodiment, the peak current is
about 60-100 amps. The absolute values of the voltages and currents
are not as important as repeatability of the voltages and currents
from spark/arc to spark/arc and avoiding ringing in the signal to
the counterelectrode 104. The amplitudes and profiles of the
voltages and currents should be as repeatable as practical.
[0074] The spark generator 200 operates under the control of the
processor. That is, the processor may specify a repetition rate, as
well as voltages, currents and/or profiles, to the spark generator
200. Alternatively, the voltages, currents and/or profiles may be
pre-configured in the spark generator 200. The spark generator 200
may be any suitable circuit, such as a switched-mode power supply
(SMPS), such as high-power thyristor or MOSFET circuit, that
produces voltages and currents as described above.
[0075] To prevent accidental exposure of a user to spark voltage,
the snout 102 may include one or more momentary contact switches,
pressure transducers or other sensors that must be activated by a
sample surface before the spark generator 200 produces a spark
signal. One embodiment of such a safety interlock system is shown
in FIG. 1. Three momentary contact switches 120, 122 and 124 are
mounted in the snout 102, such that the switches 120-124 are
activated only if the flat portion 103 of the snout 102 is fully
engaged against the surface of a sample.
Light Collection
[0076] Referring back to FIG. 5, it should be noted that the light
path 502 forms an angle 504 with the sample surface 500. FIG. 8 is
a close-up view of the analytical gap, showing a discharge region
800 between the counterelectrode 104 and the sample surface 500.
The portion of the region 800 closer to the sample surface 500 is
hotter (at about 30,000.degree. C.) than the portion of the region
800 (at about 1,500.degree. C.) closer to the counterelectrode 104.
Thus, hard line emissions 802 from elements such as phosphor,
sulfur and carbon emanate from the hotter portion of the region
800. Conversely, soft line emissions 804 from elements such as
aluminum and copper emanate from the cooler portion of the region
800.
[0077] Emissions from an analyte should be sampled from a volume of
the analytical gap where the analyte is ionized. The hard line
emissions 802 should be observed at an angle of about 1-5.degree.,
whereas the soft line emissions 804 should be observed at an angle
of about 3-15.degree.. An angle of about 3.degree. provides a good
compromise, enabling observation of both the hard line emissions
802 and the soft line emissions 804. Referring again to FIG. 5, the
angle 504 may be about 3.degree., although other small angles may
be used. In addition, multiple optical paths, possibly each at a
different angle, may be provided from the analytical gap and
recombined closer to the spectrometer 204. A mask 806 should be
used to avoid observing emissions from the hot tip of the
counterelectrode 104 or emissions from the sample surface 500.
[0078] As noted above, a combination of high-energy and low-energy
sparks and/or arcs may be used in a series of excitations,
facilitating detecting hard metals and soft metals in a sample
during different sparks/arcs. In one embodiment, the major
element(s) in the sample is(are) analyzed with a first pulse, and
trace elements in the sample are analyzed with a second pulse.
[0079] Emissions from some elements peak later during a spark/arc
than emissions from other elements. Similarly, "background"
emissions, such as from the sample surface 500 or the tip of the
counterelectrode 104, may peak earlier than emissions from some of
the elements in the sample. For example, emissions from lead peak
late, after many of the background emissions have subsided.
Time-resolved analysis of the optical signal may provide a better
signal-to-noise ratio by analyzing the spectrum for emissions from
particular elements when those emissions peak. Time-resolved
analysis of an optical signal is discussed in detail in provisional
patent application No. 60/891,320, filed Feb. 23, 2007, titled
"Time-Resolved Spectroscopy with Sensor Array" and in U.S. patent
application Ser. No. 12/035,477, by Denis Baiko, et al, titled
"Fast and Precise Time-Resolved Spectroscopy with Linear Sensor
Array," filed Feb. 22, 2008, the contents of which are incorporated
by reference as if fully set forth herein.
Spectrometer
[0080] Maintaining physical relationships, such as distances and
orientations, among optical components of a spectrometer is
necessary to maintain accuracy of the spectrometer. In a
traditional spectrometer, the optical components, such as a
structure defining an entrance slit, a diffraction grating and one
or more sensors, are rigidly mounted to a structural member made of
cast iron or Invar (FeNi), and the spectrometer is temperature
controlled to limit thermal expansion or contraction of the
structural member. Temperature control is traditionally achieved by
heating the spectrometer to a uniform and constant temperature,
although some spectrometers are cooled, rather than heated. In
either case, energy is consumed to heat or cool the spectrometer.
Sometimes electric fans are used to circulate air within or around
a spectrometer to maintain a uniform temperature. Heating a
spectrometer may necessitate selecting temperature-insensitive
sensors or cooling sensors within the spectrometer to avoid
generating heat-induced noise in the sensors.
[0081] A spectrometer in a hand-held test instrument should be
small, light-weight and consume little electrical power. FIG. 9 is
a cutaway perspective schematic diagram (with the cover removed for
clarity) of the spectrometer 204, according to one embodiment of
the present invention. Various aspects of the spectrometer 204,
including its cross-dispersed design, contribute to its compact
size, light weight and low power consumption.
[0082] Structural components, such as the case 900 and cover
(removed for clarity), of the spectrometer 204 are made of a
light-weight material, such as graphite-filled polyphenylene
sulfide (PPS), that has a small coefficient of thermal expansion
(CTE) over a range of expected ambient temperatures in contexts
where the test instrument 100 may be used. The small CTE reduces or
eliminates the need to temperature-control the spectrometer 204,
thereby conserving electrical power, while maintaining the accuracy
of the spectrometer 204. In addition, PPS is black, which assists
in absorbing stray light within the spectrometer 204. PPS may be
machined or injection molded, or a combination thereof, to produce
the structural components of the spectrometer 204.
[0083] PPS is available from Chevron Phillips, The Woodlands, Tex.,
under the tradename Ryton PPS. Polyphenylene sulfide filled with
about 40% graphite is preferred. Such a material is available under
the designation IPC-1834 from Hoerbiger America Rings &
Packing, Inc., Houston, Tex. 77023 or under the designation
"Bearing Grade" from Boedeker Plastics, Inc., Shiner, Tex. Other
polymers, high-carbon composites, glass-filled polymers or liquid
crystal polymers that exhibit or are modified, such as by filling
with carbon or another suitable filler, for a small CTE at expected
ambient temperatures may also be used.
[0084] The optical signal 210 is reflected by the mirror 202 onto
an entrance slit 901. The entrance slit 901 may be about 5 .mu.m
wide. A prism 902 vertically disperses the incoming light, as
indicated at 904. The prism 902 is located about 60 mm from the
entrance slit 901. The prism 902 is preferably made of beta alumina
or another material that is sufficiently transparent at the
wavelengths of interest.
[0085] The prism 902 is attached to a field stop (internal baffle)
906. The field stop 906 and prism 902 are shown enlarged and from
another view in the insert in FIG. 9. The prism 902 should be
mounted at its minimum deviation angle to minimize astigmatism. In
one embodiment, the prism 902 refracts light at an angle of about
6.8.degree., thus the prism 902 is tilted at about half that angle,
so the axis of the output from the prism 902 (toward the grating
910) is parallel to the floor 905 of the spectrometer case 900. The
back 907 of the field stop is angled to tilt the prism 902
appropriately.
[0086] The field stop 906 defines an approximately 1/4-inch wide
aperture 908, through which the vertically-dispersed light passes.
As noted, spark strikes on the sample surface 500 occur within a
small area, not necessarily at a single point on the surface. To
accommodate this spark "wander," the image of the spark is
defocused somewhat at the entrance slit 901; instead, the image is
focused on the internal baffle 906.
[0087] The vertically-dispersed light 904 from the prism 902
impinges on a concave holographic grating 910. The grating 910 is
about 1/2-inch thick and about 50-75 mm in diameter. The internal
baffle 906 masks off the edges of the discharge volume in the
analytical gap, thus preventing an optical signal from the tip of
the counterelectrode 104 or from the sample surface 500 from
reaching the grating 910. The grating 910 horizontally disperses
the light. The horizontally-dispersed light impinges on an array of
sensors 912. The grating 910 is constructed to have comparable
efficiencies in two different, although not necessarily
consecutive, orders. Each order may be positive or negative.
[0088] The grating 910 produces two distinct spectra, which will be
referred to as a first-order spectrum 914 and a second-order
spectrum 916, on the sensors 912. The resolution of the
second-order spectrum 916 may be greater than the resolution of the
first-order spectrum 914. Because the prism 902 vertically
disperses the incoming light 900, long and short wavelengths of the
vertically-dispersed light impinge on the grating 910 at different
angles. This angular difference causes a vertical displacement 918
between the first-order spectrum 914 and the second-order spectrum
916 on the sensors 912. In one embodiment, the vertical
displacement 918 is about 2 mm. The sensors 912 may include two
rows of sensors, one row of sensors for each spectrum 914 and 916,
although an alternative embodiment of the sensors 912 is described
below.
Order Separation
[0089] Various system parameters influence the extent of the
vertical displacement 918. If an insufficient amount of vertical
displacement 918 is provided, the two order spectra 914 and 916
partially or completely overlap each other on the sensors 912.
Depending on the heights of the sensor pixels, such an overlap may
make it impossible to achieve a clean spectrum on each set of
sensors. The upper portion of FIG. 10 schematically represents two
order spectra 1000 and 1002, as imaged on the sensors 912 (FIG. 9),
in which the two spectra 1000 and 1002 significantly overlap,
possibly preventing achieving a clean spectrum on each set of
sensors.
[0090] Returning to FIG. 9, the amount of vertical displacement 918
between the two spectra 914 and 916 on the sensors 912 depends, in
part, on the amount of dispersion 904 caused by the prism 902
which, in turn, depends on the index of refraction of the material
of the prism 902 and on the apex angle of the prism 902. However,
large apex angles result in thicker prisms, which may further
attenuate the optical signal, particularly in the ultraviolet
range.
[0091] The vertical displacement 918 also depends, in part, on the
linear magnification of the focusing system (i.e., on magnification
of the convex grating 910), on the distance 920 between the prism
902 and the grating 910 and on the distance between the grating 910
and the sensors 912. The grating 910 is a focusing element, thus
decreasing the separation 920 increases the displacement 918. In
one embodiment, the linear magnification of the convex grating 910
is about -1.
[0092] In one embodiment, the prism 902 is placed as close as
possible to the grating 910, without occluding the light path,
between the grating 910 and the sensors 912, of any portion of
either order's spectrum 914 or 916 that is of analytical interest.
Such a placement of the prism 902 in this embodiment creates a slit
901 to prism 902 distance of about 60 mm. The lower portion of FIG.
10 is a schematic diagram of two order spectra 1004 and 1006, as
imaged on the sensors 912 (FIG. 9), achieved by positioning a beta
alumina prism having an apex angle of about 8.degree. as described
above. As can be seen in lower portion of FIG. 10, the two spectra
1004 and 1006 do not overlap and provide a vertical displacement of
about 2 mm between the spectra 1004 and 1006, although the
displacement may vary with wavelength. In one embodiment, the
displacement varies from about 0.6 mm to about 3 mm over the
wavelength range of interest.
[0093] For identifying ferrous and other common metals, optical
emissions from the analytical gap that have wavelengths between
about 170 nm and about 410 nm are of interest. In one embodiment,
the grating 910 (FIG. 9) is constructed such that the efficiency of
the grating at the first order is relatively high for wavelengths
between about 247 nm and about 410 nm and relatively low outside
this range, and the efficiency of the grating at the second order
is relatively high for wavelengths between about 170 nm and about
247 nm and relatively low outside this range, although there may be
some overlap between the high-efficiency portions of the first and
second orders. The grating 910 design, the cross-dispersion
provided by the combination of the prism 902 and the grating 910
and the two rows of sensors 912 enable the spectrometer 204 to
analyze a relatively broad range of wavelengths in a relatively
small amount of space. In some embodiments, there may be a spectral
gap or overlap between the first-order spectrum 914 and the
second-order spectrum 916 on the sensors 912.
[0094] The vertical displacement 918 between the first-order
spectrum 914 and the second-order spectrum 916 may be insufficient
for two co-planar rows of sensors 912. In this case, one of the two
rows of sensors may be oriented in a plane that is perpendicular to
the sensor array 912 shown in FIG. 9, and a mirror may be used to
reflect one of the two spectra onto the perpendicular sensor array.
A side view of such an arrangement is shown schematically in FIG.
11, and a perspective view (looking slightly upward and from the
side) of such an arrangement is shown in FIG. 12. Referring to FIG.
11, a mirror 1100 reflects first order light 1102 to a
downward-facing row of sensors 1104. Second order light 1106
impinges directly, i.e. without first being reflected, on a
forward-facing row of sensors 1108. At the wavelengths of interest,
mirrors reflect longer wavelengths of light more efficiently than
shorter wavelengths. For example, below about 240 nm, about 20% of
an optical signal is lost as a result of reflecting the signal with
a mirror. The order primarily composed of wavelengths that are less
efficiently reflected by a mirror should impinge on the
forward-facing row of sensors 1108. As noted, in one embodiment,
the second order light (ranging from about 170 nm to about 250 nm)
impinges directly on the forward-facing sensors 1108. FIG. 12 shows
mounting brackets 1200 used to mount the mirror 1100 and the two
rows of sensors 1104 and 1108 to the floor 905 of the housing
900.
[0095] Spectral resolution is generally defined as the spectral
separation between the two closest peaks that a spectrometer can
resolve. For a digital sensor that includes a set of adjacent
pixels to resolve two peaks, at least one pixel between the two
peaks should receive a lower signal than its neighbors, as shown
schematically in FIG. 19. If the peaks fall on the sensor so that
the pixels with the maximum signals are next to each other, the two
peaks may not be resolved by the spectrometer/sensor
combination.
[0096] Spectrometer bandpass (BP) specifies how much spectral
bandwidth is seen at a given wavelength position. Since bandpass
limits the ability of a spectrometer to separate peaks, it is
common to refer to the BP as the spectral resolution of the
spectrometer. The BP may be calculated from the output image width
and the reciprocal linear dispersion of the dispersive element in
the spectrometer. The reciprocal linear dispersion indicates the
width of spectrum that is spread over a distance of 1 mm at the
focal plane, i.e., the sensors 912 in the description above.
Reciprocal linear dispersion, which varies with wavelength, is
given in nm/mm and is typically listed as a primary instrument
specification. The reciprocal linear dispersion of the diffraction
grating depends largely on the pitch of the grooves in the
grating.
[0097] In one embodiment of the spectrometer 204, the diffraction
grating 910 has a reciprocal linear dispersion of about 5 nm/mm,
and the entrance slit 901 has a width of about 5 .mu.m. Thus, in
one embodiment, the diffraction grating 910 provides a resolving
power of at least about 5,000 and, in another embodiment, at least
about 10,000. In one embodiment of the spectrometer 204, each
sensor 1004 and 1008 has an effective pixel pitch of about 7 .mu.m.
The resulting resolution is about 0.02 nm per pixel in the
second-order spectrum 916 and about 0.04 nm per pixel in the
first-order spectrum 914.
[0098] Each sensor 1104 and 1108 may incorporate two or more rows
of pixels in which the pixels are staggered horizontally, which
increases the effective optical resolution of the device. FIGS. 20
and 21 depict two alternate staggered pixel configurations. A
typical pixel 2000 includes a light-sensitive area 2002 and a
surrounding light-insensitive area 2004. In a typical spectroscopic
application, such as spark OES, the optics are configured to impose
a tall narrow slit image 2006 or 2100 onto the detection device. By
manufacturing the device with two or more rows of pixels, each with
a horizontal resolution of X, where the pixels in the other row(s)
are offset by a distance of X/2, the effective resolution of the
system improves to X/2. This, in turn, allows for the use of more
narrow entrance slits, which effectively improves the spectroscopic
resolution of the system.
[0099] One embodiment of the present invention operates with
uncollimated optical signals provided to the prism 902. Although
uncomimated signals may cause a small amount of aberration in the
image projected on the sensors 1104 and 1108, any "smearing" of the
image is generally in the same direction as the long dimension of
the rectangular pixels. Thus, these aberrations do not negatively
affect the resolution of the spectrometer. In addition, using lower
orders, such as first and second, of diffracted signals from the
grating 910 may minimize some aberrations.
[0100] The spectrometer described herein may be used in
applications other than hand-held analytical instruments. For
example, the spectrometer may be used in bench-top analyzers,
telescopes, telecommunications equipment, etc.
Dynamic Wavelength Calibration
[0101] In one embodiment, each row of sensors 1104 and 1108
contains about 4,096 pixels; however, in other embodiments other
numbers of pixels may be used. Sensors that include more pixels
than are necessary to image a spectrum may provide advantages. FIG.
22 is a perspective schematic diagram of a sensor 2200 and a
spectrum 2202 impinging on the sensor 2200. The sensor 2200
contains a row 2204 of sensor pixels, exemplified by pixels 2206,
2208 and 2210. If the row of pixels 2204 is just wide enough to
capture a spectrum of analytical interest, then when the sensor is
mounted in the spectrometer, the position of the sensor 2200 (or
another component, such as the grating) may need to be carefully
adjusted, so the entire spectrum is imaged, i.e., the entire image
impinges on the row of pixels 2204.
[0102] However, if the row of pixels 2204 is longer than the
spectrum 2202 is wide (as shown in FIG. 22), the sensor 2200 may be
mounted with less positional precision, as long as the entire
spectrum 2202 falls somewhere on the row of pixels 2204.
Essentially, the additional pixels, i.e., the number of pixels in
excess of the number needed to image the entire spectrum 2202,
provide a tolerance, within which the sensor 2200 may be mounted.
Once the sensor 2200 is mounted, the sensor 2200 or the processor
(not shown) may determine which pixels are illuminated by the
spectrum 2202 and, if desired, assign pixel numbers or addresses
beginning with the pixel at one end of the spectrum 2202. If the
spectrum 2202 shifts position on the sensor 2200, as indicated by
spectrum 2212, due to, for example, thermal expansion or
contraction of a component of the spectrometer or elsewhere in the
instrument, the sensor 2200 or the processor may compensate by
renumbering the pixels or reading data from a different set of
pixels, corresponding to the location where the spectrum 2212 has
shifted.
[0103] In one embodiment, the optical system and sensor is
configured such that the sensor detects a first order spectrum that
extends from about 246.9 nm to about 410 nm, and the sensor detect
a second order spectrum that extends from about 170 nm to about
246.9 nm. There should be some overlap between the two spectra at,
for example, 246.9 nm.
[0104] The row of pixels 2204 may be wavelength calibrated, i.e.,
the pixels may be associated with wavelengths, by testing a sample
that has a known composition and correlating expected peaks in the
spectrum with pixels that experience correspondingly high values of
illumination. In one embodiment, the processor automatically
wavelength calibrates the row of pixels 2204 by matching an
observed spectral feature with one of a set of stored feature
prototypes. Essentially, the processor matches the pattern of the
observed feature with a known pattern.
[0105] The pattern may include relative spacing between or among
peaks, valleys or other spectral characteristics and relative
height(s) of the peak(s), valley(s), etc. For example, FIG. 23
illustrates a known spectrum. The spectrum contains well-defined
peaks for various elements. One or more stored feature prototypes
are stored in a memory accessible by the processor. The feature
prototypes need not include information about an entire spectrum of
a material; the prototype may include information about only
selected peaks, etc.
[0106] Prototypes may be based on expected "matrix" elements in
expected samples, because these elements will likely have a strong
presence in every sample exposure. For example, for iron and steel
samples, a prototype that contains information about elemental iron
(Fe) may be used, and for aluminum alloys, a prototype that
contains information about elemental aluminum (Al) may be used.
[0107] After the instrument takes a reading, the processor searches
the data provided by the sensors for a match with one or more of
the stored prototypes. Note that the data from the sensor may
include signatures of additional materials that are included in the
tested sample, but are not represented in the prototypes. The
prototypes may be chosen so that their patterns are easily detected
among other likely materials in samples. For initial wavelength
calibration, a known standard may be used as the sample.
[0108] Once the processor identifies a prototype pattern that
matches observed data, the processor associates one or more pixels,
where one or more features of the prototype are observed, with
corresponding wavelength(s) stored with the prototype data. In one
embodiment, the processor associates a wavelength or wavelength
range to one pixel, according to an observed and matched feature,
and assigns other wavelengths or ranges to the other pixels based
on an expected linear spectral dispersion based on the geometry of
the spectrometer.
[0109] In another embodiment, the processor associates a wavelength
or range to a pixel, as described above, and calculates an actual
linear spectral dispersion observed on the sensor, based on
relative spacing between or among observed and matched features,
and associates wavelengths or ranges with other pixels, based on
the calculated linear spectral dispersion.
[0110] The wavelength calibration may create a mapping between
pixel number and wavelength. The dispersion is not necessarily
constant across the length of the detector. Thus, identifying more
peaks allows for a higher order mapping function to be used. For
example, identifying one peak allows for a 0th order "shift"
correction, two peaks allows for a 1st order linear correction, and
so on.
[0111] FIG. 24 is a flowchart that describes automatic wavelength
calibration. At 2400, a reading is taken, i.e., a spectrum from a
sample impinges on the sensors, and the processor reads at least
some of the sensors. At 2402, a search is conducted of the observed
spectrum and stored spectral feature patterns for a pattern that
matches at least a portion of the observed spectrum. At 2404, a
wavelength or range of wavelengths (hereinafter collectively
referred to in this context as a wavelength) is associated with a
pixel (such as the first pixel in the sensor array), based on a
correspondence between a second pixel that is located at the
centroid of a first observed spectral feature (such as a feature in
the matched prototype) and information (such as wavelength) about
the spectral feature pattern. This information may be stored in a
memory.
[0112] In one embodiment, other wavelengths are assigned to other
pixels based on an expected linear spectral dispersion based on the
geometry of the spectrometer.
[0113] In another embodiment, at 2406, a linear spectral dispersion
on the sensor is calculated, based on a correspondence between
another pixel at the centroid of a second observed spectral feature
and information (such as wavelength) about the second spectral
feature and the number of pixels (or distance) between the two
pixels at the centroids of the observed features. At 2408, a
wavelength is associated with other pixels, based on the wavelength
associate with the pixel at 2404 and the calculated linear spectral
dispersion.
[0114] The operations at 2404 and 2406 may, collectively, associate
wavelengths with all the pixels in the sensors. On the other hand,
these operations may associate wavelengths with only a portion of
the pixels. In that case, as indicated at 2410, necessary
operations may be repeated for other groups of pixels.
[0115] Thus, the disclosed instrument may be more easily assembled
or subsequently adjusted, without requiring high precision
positional adjustments to the sensors. In addition, the instrument
may maintain its accuracy over time and in the face of
temperature-induced dimensional changes, imperfect imaging of the
wandering spark source, mechanical vibration, physical shock and
the like by dynamically wavelength calibrating itself, based on
observed spectral features. This wavelength self-calibration may be
performed automatically at the beginning of each sample run or
after a predetermined number of runs, at other automatically
determined times (such as between sparks or after detection of a
physical shock by an accelerometer or a temperature change by a
thermistor within the instrument) or in response to a user-entered
command.
Diffraction Grating Mount Assembly
[0116] FIG. 13 shows an exploded view of a diffraction grating
assembly 1300. In the diffraction grating assembly 1300, the
diffraction grating 910 is held between a compression ring 1301 and
a diffraction grating mount 1302. Immediately behind the grating
910 is a thin (approximately 1/32-inch thick) elastomeric pad 1304,
which may be cork or another suitable material. A grating
compression plate 1306 is disposed between the elastomeric pad 1304
and the diffraction grating mount 1302. The compression ring 1301,
the compression plate 1306 and the diffraction grating mount 1302
are preferably made from the same material as the case 900 of the
spectrometer 204. The compression ring 1301 is attached to the
diffraction grating mount 1202 by two screws (one of which is shown
at 1307), which pass through holes 1308 and 1310 in the compression
ring 1301 and thread into corresponding holes (one a which is
visible at 1312) in the diffraction grating mount 1302. The
compression ring 1301 applies even pressure along the perimeter of
the diffraction grating 910, and the elastomeric pad 1304 enables
the diffraction grating 910 to expand or contract with temperature
changes without distorting the diffraction grating 910.
[0117] The diffraction grating assembly 1300 is mounted in the
diffraction grating housing 900, as shown in FIG. 14. Adjustment
screws 1400 and 1402 may be used to tilt the diffraction grating
assembly 1300. A well 1404 in the floor of the housing 900 provides
clearance for the bottom portion of the compression ring 1301.
Preferably, the diffraction grating 910 is tilted about 4.degree.
back from normal to the floor of the housing 900. Light from the
prism 902 impinges on the diffraction grating 910 at an upward
angle. Tilting back the diffraction grating 910 causes the
dispersed light to follow a path to the sensors 912 that is
approximately parallel to the floor of the housing 900.
Test Instrument Alignment
[0118] It may be necessary to align the optics of the test
instrument 100. The mirror 202 may be aligned using a setup
illustrated in FIG. 15. The counter electrode 104 is removed from
the test instrument 100 and a filler block 1500 may be inserted
from the back of the snout 102 to temporarily replace the counter
electrode assembly. A tubular mask 1502, as illustrated by front
and top news in the insert in FIG. 15, is inserted into the opening
in the front of the snout 102. An ultraviolet (UV) light source
1504 is inserted into the opening in the front of the snout 102 and
into the mask 1502, leading approximately 4 mm of the UV light
source 1504 exposed within the opening in the front of the snout
102. The tubular mask 1502 is sized to accommodate the outside
diameter of the UV light source 1504 and the inside diameter of the
opening in the front of the snout 102. Additional shielding 1506
may be used to minimize UV leakage. The mirror 202 is then adjusted
to maximize the amplitude of the signal 1508 reaching the
spectrometer 204 (not shown).
[0119] Alternatively or in addition, a visible or invisible laser
beam may be introduced into the spectrometer 204 in the vicinity of
the sensors 912 and projected backwards through the spectrometer
204 to the spark gap. A port (not shown) may be provided in the
housing 900 of the spectrometer 204 to facilitate introducing the
laser beam. Optical components in the spectrometer 204 and the
mirror 202 may be adjusted until the laser beam is detected in the
opening in the front of the snout 102 where analyte emissions are
expected to be produced. An expected path of the laser beam through
the spectrometer may be calculated, based on the wavelength of the
laser beam. It should be noted that the path taken by the laser
beam toward the diffraction grating 910 may not coincide with the
paths taken by some wavelengths of the optical signal dispersed by
the diffraction grating 910 towards the detectors 912, due to the
wavelength of the laser beam and the angle at which the diffraction
grating 910 reflects light at that wavelength. Thus, sensors 912
may be clear of the port by which the laser beam is introduced into
the spectrometer 204.
[0120] Further alignment may be performed by reflecting the laser
beam at the opening in the front of the snout 102, back along the
optical path, to the spectrometer. Alternatively or in addition, a
laser beam may be introduced into the opening in the front of the
snout 102 and directed along the optical path to the
spectrometer.
[0121] FIG. 16 is a flow chart that describes a process for
analyzing composition of a sample. At 1600, an environment in which
a portion of the sample may be analyzed is created. Creating this
environment may include purging air from the portion of the sample
that is to be analyzed. An inert gas, such as argon, may be used to
purge the air.
[0122] At 1602, a portion of the sample is excited. The sample may
be excited with an electric spark/arc, a laser, glow discharge or
another suitable mechanism. If an electric spark/arc is used, a
spark gap is created between a counterelectrode and the sample. The
counterelectrode and the sample are electrically connected to a
spark source, which produces a suitable potential having an
appropriate waveform. A potential difference between the
counterelectrode and the sample breaks down the gas in an
analytical gap and erodes a portion of the sample into the
analytical gap. The potential may be reduced, and current may be
increased, to ionize the sample material in the analytical gap. The
ionize sample material emits an optical signal.
[0123] At 1604, the optical signal is collected. The optical signal
is routed, via an optical path, to a spectrometer. At 1606, the
optical signal is wavelength-dispersed. The optical signal may be
cross-dispersed. At 1608, the intensity of the dispersed optical
signal is measured at wavelengths of interest. One or more arrays
of sensors may be used to measure the intensities of the dispersed
optical signal. If the optical signal is cross-dispersed, one set
of the sensors may be disposed a distance away from the other of
the sets of sensors, according to the amount of cross-dispersion.
At 1610, the intensity measurements are processed to determine the
composition of the sample. The processing may be performed by a
processor executing instructions stored in a memory. The process
may be repeated, as indicated 1612, for a series of measurements.
Data from the series of measurements may be averaged and/or
parameters of the excitation (1602) may be varied for each of the
repetitions.
[0124] FIG. 17 is a block diagram of major components of the test
instrument 100. Instructions for a processor 1700, as well as
spectral feature prototypes, may be stored in a memory 1702.
Analytical results from samples may also be stored in the memory
1702 and displayed on the touchscreen 110 and/or provided to an
external device via a wired or wireless data port 1704. In
addition, the memory 1702 may store tables of compositions of known
materials (such as alloys) for comparison to compositions of test
samples, and results of this comparison may be displayed on the
screen 110 and/or provided via the port 1704.
[0125] Referring to FIG. 1, the touchscreen 110 is readable while
the test instrument 100 is in most orientations. However, in some
cases, the touchscreen may be difficult to read. A hinged (tilt-up)
screen may be used in some OES, x-ray fluorescence (XRF) or other
of hand-held, self-contained test instruments. One embodiment of
such a tilt-up screen is shown at 1800 in FIG. 18. A flexible
ribbon cable or other suitable flexible wire is used to connect the
screen 1800 to the processor or other circuitry within the test
instrument 100.
[0126] Although a spectrometer having a wavelength range of about
170 nm to about 410 nm has been described, spectrometers according
to the present invention may have other wavelength ranges.
[0127] A hand-held, self-contained, battery-powered test instrument
has been described as including a processor controlled by
instructions stored in a memory. The memory may be random access
memory (RAM), read-only memory (ROM), flash memory or any other
memory, or combination thereof, suitable for storing control
software or other instructions and data. Some of the functions
performed by the test instrument have been described with reference
to flowcharts. Those skilled in the art should readily appreciate
that functions, operations, decisions, etc. of all or a portion of
each block, or a combination of blocks, of the flowcharts may be
implemented as computer program instructions, software, hardware,
firmware or combinations thereof. Those skilled in the art should
also readily appreciate that instructions or programs defining the
functions of the present invention may be delivered to a processor
in many forms, including, but not limited to, information
permanently stored on non-writable storage media (e.g. read-only
memory devices within a computer, such as ROM, or devices readable
by a computer I/O attachment, such as CD-ROM or DVD disks),
information alterably stored on writable storage media (e.g. floppy
disks, removable flash memory and hard drives) or information
conveyed to a computer through communication media, including wired
or wireless computer networks. In addition, while the invention may
be embodied in software, the functions necessary to implement the
invention may alternatively be embodied in part or in whole using
firmware and/or hardware components, such as combinatorial logic,
Application Specific Integrated Circuits (ASICs),
Field-Programmable Gate Arrays (FPGAs) or other hardware or some
combination of hardware, software and/or firmware components.
[0128] While the invention is described through the above-described
exemplary embodiments, it will be understood by those of ordinary
skill in the art that modifications to, and variations of, the
illustrated embodiments may be made without departing from the
inventive concepts disclosed herein. Furthermore, disclosed
aspects, or portions of these aspects, may be combined in ways not
listed above. For example, the spectrometer described above may be
used in other contexts, such as terrestrial or extraterrestrial
astronomy, including in combination or within telescopes and
satellites. Accordingly, the invention should not be viewed as
limited.
* * * * *