Laser-based in-situ exhaust gas sensor

Abstract

A laser absorption spectroscopy exhaust gas sensor comprising: an optical cell having a porous wall [404] having holes with an average diameter in the range of 0.1 nm to 1 mm; a gold mirror [405] within the optical cell positioned to support a multi-pass optical path [412] within the optical cell; an active heating element adapted to heat the optical cell to prevent condensation; a laser adapted to generate a laser beam; an optical detector [408] adapted to detect a returning laser beam; and a processor for controlling the laser and the active heating element and for analyzing signals from the optical detector to identify a gas in the optical cell.

Description

Technical Field

[0001] The present invention generally relates to methods and apparatus for detecting the concentration of gases and more particularly to techniques for in-situ detection of low concentration components at elevated temperatures using laser absorption spectroscopy in the presence of particulates and moisture, such as vehicle exhaust and power plants. Background Art

[0002] Laser-based sensing technology has been widely used in applications ranging from environmental monitoring to healthcare. In these sensors, an optical cell is filled with a mixture of gases being analyzed and a laser enters the cell through a window of the cell and exits the cell through a different or the same window after traveling a distance through the analyzed gas inside the cell. Sometimes, to increase the sensitivity of the sensor, the distance traveled through the mixture is increased by using one or more reflectors. Most of these highly explored areas are in clean environments where there is no risk of the laser beam being blocked by particles or condensation. However, one of the important advantages of laser-based technology is the ability to monitor gases at high temperatures and in harsh environments, such as measuring combustion product concentrations. However, one of the problems with existing devices is the problem of keeping the optical cell clean throughout the working life of these devices. After long-term exposure in these environments, particles and moisture deposit on and / or corrode the optical reflective or transparent surfaces, causing these instruments to malfunction. Summary of the invention

[0003] In one aspect, the present invention provides a gas sensor comprising an actively heated cell having walls that are porous to the gas but at the same time prevent most particles from entering the cell, thereby keeping the cell clean. The porous wall design exploits knowledge of the local particle distribution. Too small a pore size disadvantageously limits the diffusion / flow rate of the sample. In this regard, the pore size is selected as an average pore size that is constrained to be equal to the size of the smallest particle expected by the probability cutoff in the flow. For example, in the case of diesel automobile exhaust, the particle distribution falls into the following broad categories:

[0004] Coarse particles or PM10 – particles with an aerodynamic diameter ≤ 10 μm

[0005] Fine particles or PM2.5 – particles with a diameter of ≤2.5 μm

[0006] Ultrafine particles, with diameters below 0.1μm or 100nm

[0007] Nanoparticles, characterized by diameters less than 50 nm

[0008] As the sample cools and hot particles from combustion nucleate, nucleation mode particles begin to appear more. If these sensors are carefully placed in a location where the sample is still hot or has not had enough time to nucleate (~450-800°C), this process may occur primarily after the location of these sensors. In this case, we only need to worry about larger particles escaping from the diesel particulate filter placed upstream. In this specific case, the pore design will be designed for particles larger than ultrafine particles (>100nm) and a characteristic pore size of 100nm will be utilized. But conversely, if the sample is expected to contain ultrafine particles, but the smallest nanoparticles are estimated to be mostly trapped upstream or too small to affect the reflectivity of the mirror during the service life of the sensor, then a finer pore size (~20nm) may be considered necessary. One way to achieve effective particle removal is to design a filter element with ultrathin walls with small pockets. This idea is implemented in a proof-of-concept prototype discussed later and is called the "gills" of the filter.

[0009] The actively heated cell is maintained at a temperature above the saturation temperature of all components to prevent condensation. Typical temperatures that can be maintained in automotive exhaust are in the range of 300-600°C. The heated device can also be equipped with a mirror that is partially reflective or has any number of holes in the reflective surface. The "holes" may or may not be physical holes and may simply be transparent areas in an otherwise reflective surface. Depending on the temperature distribution in the cell, the transmissive or reflective portion should contain all appropriate materials required to transmit or reflect the laser within tolerances under design conditions.

[0010] Thus, in one aspect, the present invention provides a laser-based in-situ exhaust gas sensor comprising: a) a laser adapted to generate a laser beam at an operating wavelength; b) an actively heated cell comprising an optical reflective element adapted to produce a beam path through the cell; c) a laser beam waveguide adapted to deliver the laser beam from the laser to the actively heated cell, wherein the laser is thermally isolated from the actively heated cell; d) a laser beam detector adapted to detect the laser beam after it has passed through the actively heated cell, and e) a signal processor adapted to identify gases in the cell based on an absorption spectrum; wherein the actively heated porous cell is in contact with a heating element adapted to maintain the actively heated cell at a temperature above the saturation temperature of components in the exhaust flow to prevent their condensation; wherein the actively heated porous cell has a wall having pores having an average diameter large enough to allow particles desired to be sensed to pass through the pores and small enough to block undesired contaminant particles from passing through the pores, thereby mitigating undesired effects of contaminants on laser sensing inside the cell.

[0011] In some embodiments, the cell has a multi-pass laser beam pattern due to multiple reflections between multiple mirrors under the conditions of the cell. The reflectivity of the mirrors is selected to reflect the laser beam with an attenuation below that required to detect the output beam intensity at a laser detector with an SNR>100. In some embodiments, a protective top layer of silicon dioxide and Al2O3 can be applied over the gold or other metal mirrors.

[0012] In one aspect, the present invention provides a laser absorption spectroscopy exhaust sensor comprising: an optical cell having a porous wall having pores with an average diameter in the range of 0.1 nm to 1 mm; a gold mirror within the optical cell positioned to support a multi-pass optical path within the optical cell; an active heating element adapted to heat the optical cell to prevent condensation; a laser adapted to generate a laser beam; an optical detector adapted to detect a returning laser beam; and a processor isolated from the optical cell by a flexible conduit, the processor being used to control the laser and the active heating element and to analyze signals from the optical detector to identify a gas in the optical cell.

[0013] The laser absorption spectroscopy exhaust gas sensor also includes an optical waveguide adapted to carry the laser beam from the laser to the optical cell and a return laser beam from the optical cell to the detector, and adapted to thermally isolate the laser from the optical cell.

[0014] Preferably, the porous wall is composed of porous sintered metal powder, wire mesh or ceramic. Preferably, the porous wall has pores with an average diameter large enough to allow the gas desired to be sensed to pass through the pores and small enough to block undesirable contaminant particles from passing through the pores. Preferably, the pores have an average diameter in the range of 10 nm to 10 microns.

[0015] The laser absorption spectroscopy exhaust sensor may include a protective top layer on the gold mirror.

[0016] The active heating element may be integrated into the porous wall or may be attached to the porous wall.Preferably, the active heating element is adapted to heat the optical cell to a temperature above the saturation temperature of components in the exhaust gas to prevent condensation.

[0017] Preferably, the walls of the optical cell have non-porous portions.

[0018] Gold mirrors consist of unprotected gold deposited on a substrate. Gold mirrors include concave mirrors.

[0019] The gold mirror may include opposing spherical mirrors, the opposing spherical mirrors having different focal lengths.

[0020] The porous wall has a plurality of portions having different wall thicknesses, the plurality of portions being adapted to enhance gas flow into the optical cell.

[0021] The multi-pass optical path has a length greater than 50 cm and a physical length less than 5 cm.

[0022] The multi-pass optical path has partially overlapping spot patterns on the mirror away from the entrance aperture and the exit aperture, the spot patterns having an elliptical shape on the mirror.

[0023] The sensor may further include fastening threads positioned around the optical cell and adapted to secure the sensor to a vehicle exhaust duct. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 : An overview of the laser-based in-situ exhaust gas sensor architecture.

[0025] Figure 2 : Cross-sectional view of the thermal part of the sensor variant.

[0026] Figure 3 : Figure 2 Axial view of the hot section of the sensor variant shown.

[0027] Figure 4 : Figure 2 3D cross-section of the thermal section of the sensor variant shown.

[0028] Figure 5 : Cross-section of the hot section of the porous sleeve variant.

[0029] Figure 6 : Exploded view of the thermal section of the porous sleeve variant.

[0030] Figure 7 : Cross-section of the thermal section of the porous window variant.

[0031] Figure 8 : Exploded view of the thermal section of the porous window variant.

[0032] Fig. 9 : Cross-sectional view of a simplified tunable optical design to capture the beam on the detector.

[0033] Fig.10 : Side view of the simplified tunable optical design to capture the beam on the detector.

[0034] Fig.11 : 3D view of the simplified tunable optical design capturing the beam on the detector.

[0035] Fig.12 : Cross-sectional view of a simplified tunable optical design for trapping a light beam on a fiber / bundle / waveguide.

[0036] Fig.13 : Cross-section of a simplified fixed optical design for trapping a light beam on a fiber / bundle / waveguide.

[0037] Fig.14 : Cross-section of a simplified fixed optical design to capture the beam on the detector.

[0038] Fig.15 : A porous optical cell with spiral heating elements.

[0039] Fig.16 : A porous optical cell with straight heating elements.

[0040] Fig.17 : A hybrid porous optical cell design concept that combines a single-component construction of the porous window and the cell body without the use of end caps.

[0041] Fig.18 : Cross-section of the first generation exhaust gas sensor prototype.

[0042] Fig.19A , Fig.19B , Fig.19C : View of the porous wall implemented in the first generation prototype.

[0043] Fig. 20A , Fig. 20B : Detailed view of the coating on the entrance mirror and substrate.

[0044] Fig.21 : An arrangement of unequal focal length multi-pass cells showing a conical spot pattern.

[0045] Fig. 22 : Partially overlapping spot patterns obtained for the ultra-compact multi-pass cell configuration.

[0046] Fig.23 : Example non-elliptical beam pattern. DETAILED DESCRIPTION

[0047] One of the first explored applications for this sensor design was automotive exhaust monitoring. Figure 1 Describes this use case. For example, Figure 1The portion of the pipe 101 in represents any pipe that transports gas. However, specifically in the case of vehicle exhaust, it can represent a portion of an exhaust pipe. The arrow illustrates the direction of gas flow through the portion. The high temperature porous optical cell (POC) 102 is the central element of the present invention having a porous wall. The pore size can be selected based on knowledge of the average particle distribution in the environment. In the probability of particle size relative to the probability distribution function, the average pore size can be equal to the smallest particle in the flow below the design threshold or the fraction of the smallest particles in the flow below the design threshold. Typical sizes of acceptable holes in automobile exhaust pipes are 10 nanometers - 10 microns. More generally, the average pore size can be selected to be a diameter ranging from molecular sieves (~0.1nm) to the size of a simple wire mesh ~1mm.

[0048] Although shown as a cylindrical element in this figure, the surface can be any arbitrary shape, including other cylindrical and prismatic shapes. The fastening portion 103 is used to mount the optical cell within the pipe 101. Although 103 is shown here as a threaded joint, it can be any other leak-proof joint, including a flange seal (as in the case of the first prototype) or a thread. In the illustrated example, a typical design of oxygen / nitrogen oxide sensor mounting is used to enable simple after-sales replacement. The head of the sensor 104 is called the end cap of the sensor and exists outside the scope of the pipe 101. This can be used to tighten the gas seal with the exhaust pipe and also provide space to thermally isolate the high temperature optics (HTO) chamber 105 (so named because it houses the high temperature optics to emit laser light and receive it back after the laser light has left the optical cell 102 containing the sample gas). The temperature of the HTO chamber is closely related to the heat transfer from the exhaust pipe 101. In our first prototype, the maximum measured temperature in the HTO chamber was always below 150°C. Conduit 106 (flexible or rigid) is used to transport and protect optical fibers and wires that carry light, voltage, current, or data to and from sensor electronics box 107, which can house portions of the sensor that need to or can be isolated from high temperature areas of the assembly. Several important components of electronics box 107 can include a) lasers, b) laser control hardware, c) photodiodes, d) analog-to-digital converters or digital-to-analog converters, and / or e) processing computers with communication units. Several of the most commonly used operating modes of laser control boxes are described below. The laser temperature controller unit can maintain a fixed temperature setting of a tunable laser diode or chip or array via a thermoelectric cooler (TEC) inside the laser housing. Additional external cooling (e.g., water or fin-based passive cooling) may be required to maintain the target laser temperature. The laser current controller is used to maintain a fixed laser current or modulate the laser current and is a mechanism to simultaneously control the laser wavelength and intensity. Current tuning is the most typical mechanism to generate the necessary patterns in the laser output wavelength / frequency and intensity. This pattern is then used in direct absorption, wavelength modulation, frequency modulation, or other established absorption spectroscopy techniques. This approach can also be used with broadband absorption techniques without using any scanning mode. The output laser beam after absorption in the POC can be captured on a detector directly or by using a waveguide. The resulting detector voltage or current signal is then digitized by an analog-to-digital converter and then interpreted in an embedded computer. Another version might use fully analog signal generation and an analog phase-locked filter to obtain final data that can be used with common absorption spectroscopy techniques. Given the relatively noise-free surroundings and the temperature ratings of the components, photoacoustic spectroscopy can also be used with the current setup with necessary modifications to the architecture (such as the use of microphones and tuning forks in the POC), with some modifications.

[0049] Figure 2 (section perpendicular to the pipe axis), Figure 3 (view along the axis) and Figure 4 (3D cross-section) provides a more detailed view of the overall design architecture of the "hot section". The design elements in these figures are one of the variations of this architecture. Other variations are described in the following figures. The optical fiber (201, 301 and 410) delivers the laser beam from the laser housed in the electronics box 107. The optical fiber material preferably has the following properties: a) it can physically withstand the temperatures to which it is exposed, b) it can transmit the laser wavelength with an acceptable amount of loss ("acceptability" can be achieved through various modifications in the design as explained below) at the highest design temperature distribution, and c) it can be single-mode, multi-mode fiber or hollow core waveguide. The "acceptability" of fiber transmission comes down to the ultimate detectability of the output laser beam. There are many parameters that can affect this. For example: 1. Detection sensitivity: At near-infrared wavelengths (~1.6μm), the detector sensitivity is very high and very lossy fibers may be acceptable, while at 10μm, the detector sensitivity is reduced and even low-loss fibers may be unacceptable; 2. Number of reflections and multi-mirror reflectivity: Each reflection reduces the intensity of the output beam. Therefore, the design can choose fewer reflections to increase the output beam power and make the fiber transmission work. 3. Fiber length: The loss / fiber length may be high, but the fiber length can be shortened to reduce the amount of laser intensity loss. An example of a fiber material that can withstand high temperatures (>2000°C) and can transmit lasers with wavelengths greater than 4μm is sapphire. Fluoride glass fibers can transmit a longer wavelength range, but can withstand a smaller temperature range.

[0050] The laser beam is launched into the high temperature optical cell 218 using the fiber collimation system 210 and 211 and through the laser transparent window 213, which is made of a material that can withstand the design temperature and moisture exposure and can transmit the wavelength of interest. For example, a calcium fluoride (CaF2) window can transmit 0.18–9.2μm with a transmittance of >60% and withstand >1000°C, but it will begin to react with moisture at temperatures >300°C. Therefore, careful design is required to reduce the temperature at the location where the CaF2 lens and window must be used. Considering that the transmission range over the temperature range is satisfactory, a material such as sapphire can be selected to withstand the hotter and humid environment of the window 213 material. Additional improvements in transmittance and environmental resistance can also be achieved by applying a thin film coating on the exposed surface. As mentioned earlier, the back side of the window 219 can be a reflective polished metal or can be coated with a dielectric layer and have the desired curvature with the "hole". A similar configuration can also include a reflector and a window as two separate components placed back to back (implemented in the first prototype). The window surface may be angled relative to the other faces to reduce etalon effects or edge effects sometimes observed due to parallelism of the window faces. Figure 2 202, 204, 205, 212, Figure 3 311, 312, 313 in the HTO chamber) to adjust and fix the laser alignment. After entering the pool, the laser beam bounces between the reflectors 213 and 216 and eventually leaves the pool through the hole in the reflector 213. The laser beam is then collected by the focusing mirror 203 / 302 onto the photodetector 206. The HTO chamber base 207 / 304 is thermally insulated from the sensor end cap 214 / 310 by a thermally insulating gasket / layer 208 or 305. The thickness and material of the gasket / layer 208 / 305 are selected to control the temperature distribution in the HTO chamber. The temperature distribution of the HTO chamber-gasket / layer 208 / 305 depends primarily on the thermal conductivity through the gasket, the temperature at the end cap, other possible heat dissipation modes of the HTO chamber, and the ambient temperature. Additional passive / active heat dissipation can be added inside and outside the HTO chamber 105 to enhance the cooling of the HTO chamber.

[0051] The exhaust gas flows through the pipe section (209 / 307 / 401) in the direction of the arrow. Except around POC 308, Figure 3The airflow in is perpendicular to the paper. POC 308 is "immersed" in the high temperature flow of exhaust gas. The sensor is fixed in the pipe using fastening threads 309 / 406. Please note that the thread can be an M20x1.5 thread commonly used in exhaust NOx sensors as a simple replacement for currently used vehicle exhaust sensors. As shown by small arrow 402, the filtered exhaust gas flows through the porous wall 215 or 404 and pours into or diffuses into the pool 218. As shown by small arrow 403, the flow diffuses out of the pool. The direction of these arrows is for illustrative purposes only and should not be confused with the actual dynamics of the flow. The construction of the porous wall 215 / 404 is possible by various methods including the additional methods described in the following sections. One construction method is to make a cylindrical pool with a porous sintered metal (e.g., austenitic nickel-chromium-based super heat-resistant alloy, steel, etc.) powder or wire mesh with appropriate porosity. Another variant may include a porous ceramic wall structure (e.g., silica, alumina, etc.).

[0052] like Figure 3 As shown, the material of the pool wall is actively heated using a heating wire, element or band 306 to avoid condensation, and in some embodiments, it can also be temperature controlled by using embedded (multiple) thermocouples or other temperature monitoring devices. Whether the temperature is kept constant by active temperature control depends on the use case. Active temperature control can be used to always match the external sample temperature in the pipeline. Sometimes, the temperature in the exhaust pipe is much higher than saturation. Therefore, it only needs to be maintained above the threshold. In these cases, uncontrolled temperature can be advantageous because it can retain temperature-dependent components such as NOx. But in this case, the temperature change needs to be measured, characterized and included in the signal interpretation model. After testing involving the first prototype, it was concluded that at the beginning of operation, the temperature of the POC pool only needs to exceed the saturation temperature of the exhaust gas (usually less than 60°C). The sensor is quickly heated by a large amount of heat from the exhaust flow, and there is no need to maintain a certain temperature. The operation of the sensor itself does not depend on the temperature of the flow. Compared with traditional electrochemical sensors that need to be preheated to 600°C, this also significantly shortens the "ignition" time.

[0053] An example laser beam propagation path through the sensor can be shown by the black dashed line 412. The laser beam is emitted from the optical fiber into free space through the collimating optical device (shown here as a lens; a reflector can also be used) 411. The laser beam then enters the POC through the window 407 and the "hole" in the reflector or through a partially transmissive mirror. After entering the pool, it begins to reflect back and forth between the input / output mirror 217 and the back reflector surface 219 or 405. The pattern formed by multiple reflections between these surfaces determines the optical path length of absorption. An example laser multi-pass pattern is the Herriott or White multi-pass architecture, which is widely used in the practice of laser sensing. Any other alternative pattern can also be achieved by using, for example, a diffuser or a spherical mirror. The beam can leave the same or different holes in the input / output mirror 217. The output beam can be collected by various methods to collect the output beam onto a photodiode (303 or 408), which may or may not include lenses, mirrors, optical fibers, fiber bundles, waveguides, or a combination thereof, as described below or in subsequent sections. This multi-pass arrangement is used to obtain an optical path length greater than 50 cm within a physical length of 5 cm. The light beam can be reflected by using a concave mirror or a plane mirror 409 on the surface of the photodiode 408. The photodiode converts the incident light beam into a current or voltage, which is then transmitted to an amplifier and then to an analog-to-digital converter. The received signal is then analyzed to obtain information about the concentration and / or temperature of the mixture, which is usually practiced in laser absorption spectroscopy by methods such as direct absorption spectroscopy or wavelength modulation spectroscopy (WMS).

[0054] Although the above embodiments have been described for illustration using optical wavelengths of laser beams, the principles of the invention generalize to wavelengths outside the optical range, from microwaves to X-rays. In general, the upper limit of the beam size can be limited to the maximum physical size of the POC diameter. Examples of beam diameters range from 0.1 microns to 10 centimeters. Window and mirror materials and the beam launch method (fiber / non-fiber) may need to be appropriately modified depending on the wavelength used. In general, at any wavelength, the core spectroscopic technique used is absorption spectroscopy. However, the exact measurement scheme depends on the tuning mechanism, if tuning is possible or even required. For example, wavelength modulation spectroscopy can only be used for current tunable lasers. Direct absorption can be used for non-tunable and tunable lasers, broadband and narrowband lasers, but it does have a series of disadvantages.

[0055] The sensor can be made by implementing a combination of the following design alternatives to various parts of the previously discussed designs:

[0056] A. Alternative Designs for Porous Wall Construction

[0057] Constructing the POC 102 as a single component may affect its manufacturability and quality assurance. As shown, the POC is connected to the sensor end cap 501 / 701. Figure 5 , Figure 6 , Figure 7 and Figure 8 A pair of alternative approaches are shown in .

[0058] 1.Porous ceramic / metal casing structure ( Figure 5 and Figure 6 ):

[0059] In this design, the POC is divided into 3 parts - a) a rigid tubular porous or non-porous part 602, b) a porous sleeve (metal or ceramic) 502 / 604, and a bottom cover 504 / 606 that houses the back reflector, which is fixed in place with epoxy, a layer of high temperature putty, or any other high temperature adhesive (503). Although these parts are shown here as being attached to each other by threads (508 / 601, 506 / 603, and 605), they can be fastened by flanges or even high temperature adhesives or putties. Once they are assembled ( Figure 5 ), a bead of high temperature epoxy or putty or any other high temperature adhesive may be used, or a metal sleeve may be welded to seal the joint (505). An alternative heating mode is also shown by 507, where a strand of heating wire / ribbon runs along the interior of the wall of 602. In this context, the term "high temperature" may mean warm (~200C), moderate (exhaust-like temperature ~500C), or fire-like (~1500C), depending on the situation.

[0060] 2.Porous ceramic / metal window structure ( Figure 7 and Figure 8 ):

[0061] Similar to the sleeve design, the POC is divided into 3 parts - a) a rigid tubular porous or non-porous part with multiple "window"-like openings 802 and threaded parts (706 / 804, 708 / 801 and 805) for assembly, b) two or more porous windows (metal or ceramic) 702 or 803, which can be bonded or fixed to 802 via other leak-proof fastening methods (for example, it can also be welded to the rest of the POC body), and c) a bottom cover 705 / 806 that houses the back reflector 704 attached by an adhesive layer 703. The heating architecture (707) shown here is similar to the one shown before, except that the density is a little higher than before. These porous shapes can be manufactured by 3D printing or casting and then sintering methods to achieve the desired shape.

[0062] B. Alternative Designs for High Temperature Optics (HTO) Chambers

[0063] The emission and collection of the beam into and out of the POC can be accomplished by a number of other methods listed below ( Fig. 9 , Fig.10 , Fig.11 , Fig.12 , Fig.13 and Fig.14 ):

[0064] 1. Simplified adjustable optical design ( Fig. 9 , Fig.10 , Fig.11 and Fig.12 ):

[0065] Fig. 9 , Fig.10 , Fig.11 and Fig.12 A pair of variations of the previously discussed design are shown in Figure 1 that allow for greater compactness at the expense of reduced adjustability. The difference between the two variations is that the laser beam after leaving the optical cell through the hole in the mirror (904 / 1208) can be captured either a) directly on the detector 901 or 1102 ( Fig. 9 , Fig.10 and Fig.11 ) or b) to a detector by first capturing it on an optical fiber, optical fiber bundle or any other waveguide 1204 and then transmitting to a photodiode, in which case the photodiode should be housed in the electronics box 107. The optical fiber 1204 may have a thermally insulating sleeve as shown at 1205. The focal length and distance of the capture lens (1207) are selected so as to confine most of the captured beam shape to the numerical aperture (NA) at the collecting fiber tip 1206. Common elements of this design include a) tilt adjustment elements (906 or 1003) for the laser fiber (907 / 1201) launch / collimation components (905, 1101, 1202, and 1203), b) HTO chamber substrates 903, 1004, and 1105, c) focused output beam capture lens (with or without anti-reflection (AR) coating) 902, d) planar displacement (mounted XY translation) beam capture optomechanical elements (1001 / 1103 and 1002 / 1104), and e) sensor end cap 1106 mounted on a tube section 1107. Although the insulating gasket / layer 208 or 305 is not shown, this design may be used with or without the insulating gasket / layer 208 or 305.

[0066] 2. Simplified fixed optical design ( Fig.13 and Fig.14 ):

[0067] All previous designs are better suited for small-scale manufacturing processes with greater design flexibility. Once a specific design is confirmed and ready for scale-up, these variants remove all adjustability to increase manufacturing cost-effectiveness and throughput. Fig.13 The version shown in FIG. 1 includes a) a receiving fiber, bundle or other waveguide 1301, b) a fiber protective sheath 1302, c) a beam capture lens 1304, d) an HTO chamber substrate 1303, e) an insulating gasket / layer 1307 between the substrate and the end cap 1305, f) a fiber launch / collimation optic 1309, and g) an input fiber 1310. The grooves used to house the optics (e.g., 1308) in the HTO chamber substrate are mounted with precision grooves that are at specific angles and locations in 3D space within the tolerances required for the laser beam pattern in the POC with a given hole location in the mirror 1306. Fig.14 A variation is shown in which the captured light is captured directly on the detector 1401. However, in this approach, additional concerns regarding heat dissipation from the photodiode through active or passive cooling strategies need to be addressed. In an alternative variation, the laser can also be mounted directly on the substrate similar to the detector if adequate cooling can be provided.

[0068] C. Alternative Heating Arrangements for Porous Optical Cells (POCs)

[0069] To prevent condensation of components near their saturated vapor pressure in the flow (1503) along the tube section 1504, the temperature of the POC 1501 / 1602 is raised by heating as previously described. In addition to the configuration discussed, a spiral ( Fig.15 ) or straight line ( Fig.16 ) pattern of heating elements 1502 or 1601 (eg, ribbons, wires, etc.). In one manufacturing mode, the heating elements can be held in the mold during the porous wall manufacturing process (eg, sintering or gel-assisted solid phase reaction process).

[0070] And finally, it must be clarified that any of these sub-design elements can be combined with other elements to form a hybrid concept. Fig.17 As shown, the porous window design element 1701 is combined with the original single element design 1702 without the use of the bottom cover 806. In this hybrid concept, the back reflector 1703 is bonded to the bottom of the POC through the holes with a high temperature bonding material such as putty or ceramic adhesive.

[0071] This type of in-situ exhaust gas sensor has been successfully verified in the exhaust gas of diesel engines. Fig.18The design of the sensor is illustrated. This is a cross-sectional view of the prototype when installed in the exhaust duct 1801. The prototype uses a flange pair (1802 and 1803) and a spiral wound gasket 1807 to achieve a gas seal between the HTO chamber 1805 and the exhaust duct 1801. The seal between the porous optical cell (POC) 1810 and the HTO chamber 1805 is achieved by a high temperature RTV or epoxy seal between the window socket 1808 and the wedge-shaped cylindrical window placed therein. The mirrors in the POC cell are represented by 1809 (top) and 1812 (bottom). The effective optical path length of this design is 1.37m. The top reflector 1809 is epoxy-connected to the lowest pocket in the HTO flange 1802. The bottom reflector 1812 is epoxy-connected to the bottom mirror cover 1813. The bottom mirror cover is fixed at a distance from the bottom surface of the HTO flange by using a stainless steel rod 1811. High temperature wires and thermocouples are fed through peripheral grooves 1804 and sealed using high temperature RTV sealant. Tapped holes 1806 on HTO flange 1802 are used to mount high temperature optical components.

[0072] The heated porous wall of the POC cell is a key element of the design and is Fig.19A , Fig.19B and Fig.19C This consists of three parts - a silica-based 3D printed ceramic porous wall 1901 and stainless steel end flanges 1902 and 1903. The porosity of the wall 1901 is controlled by the sintering temperature and time of the unfired (green) 3D printed wall. The 3D printed structure initially consists of plastic and ceramic powders. After the firing (sintering) process, the structure is made entirely of ceramic and is therefore able to withstand temperatures far above typical vehicle exhaust temperatures (up to 600°C). During installation, the end flanges (1902 and 1903) can be slid in and out of grooves in the ceramic wall 1906. The pockets on the wall 1904 are known as the "gills" of the POC cell, enhancing the flow of ambient gas into the cell and resulting in ultra-low residence time in the cell. The ridges 1905 on the POC wall are used to mount wall heaters of nickel-chrome or Kanthal wire, as shown in 1907. These heating elements are used to preheat the cell to above saturation temperature (usually <60°C) at start-up. Once exhaust gas begins to flow, the exhaust gas will quickly heat the POC cell to a temperature >150°C, leaving no room for condensation to occur and thus requiring active heating of the cell.

[0073] Mirror coatings for high temperatures

[0074] Typical mirrors used in multichannel cells are shown in Fig. 20A , 20B2001 in FIG. The mirror may or may not contain holes (e.g., 2002), depending on the desired path length required by the design. Reference 2003 shows a cross-sectional view of the mirror (section AA). Reference 2004 shows an enlarged view of the layers in the mirror coating. The shaded areas 2005, 2006, and 2007 represent the protective top layer, the reflective layer, and the adhesive layer, respectively. 2008 refers to the polished substrate on which the mirror coating is applied. It has been experimentally observed that the commonly used protective top layer (2005) on gold has an adverse effect on the stability of the coating at high temperatures >400°C. This behavior may be due to the uneven expansion of the layer causing stress enhancement leading to cracking of the reflective coating. Another candidate protective top layer, HfO2, was found to significantly change transparency at high temperatures. These behaviors are rather non-intuitive and are a major new discovery of our research. The only reflective mirror coating that survived our high temperature exposure tests was the unprotected gold. The integrity and thickness of the gold coating are also important for estimating its strength at higher temperatures. Furthermore, a bonding layer, typically a Cr, Ti or Ni layer, sandwiched between the gold coating and the substrate (fused quartz in our test case) is observed. However, future research will be conducted to find a suitable protective top layer that can survive the high temperatures. Furthermore, different substrate materials such as stainless steel (particularly SS 310 due to its close match of thermal expansion coefficient to gold), nickel alloys or titanium can also be implemented.

[0075] Multi-pass pool design

[0076] The multi-pass optical cell design we implemented is based on two new concepts: unequal focal length spherical mirrors and partially overlapping spot arrangements.

[0077] a) Unequal focal length: by Fig.21 The concept of different focal lengths is illustrated. Most of the Herriot-type multi-pass cells mentioned before rely on spherical concave mirrors of equal focal length or one concave and one flat mirror. Here, we extend the concept to unequal focal length mirrors facing each other. In the figure, the entrance mirror is represented by the mark 2101. At least two holes are machined on the mirror, as shown by black dots 2103. Another mirror represented by 2102 is called the return mirror. The light spots formed on the mirror due to multiple reflections are shown by white spots 2104. The smaller the size of the mark selected to represent these light spots, the greater the number of reflections experienced before reaching the light spot. The mirror has a longer focal length than 2101, which causes the pattern on it to be relatively dense, as shown in the computer-generated light trajectory. This is only advantageous due to our next concept.

[0078] b) Partially overlapping spot arrangement: Conventional multi-pass cell design requires that the entire pattern formed on the mirror must not have any overlap with its neighboring patterns to prevent optical fringing or etalons. However, this is a stricter requirement than is actually required. To prevent fringing, only the entrance and exit beams must not overlap with their neighbors. We introduced this new concept into our multi-pass cell design and achieved a radically reduced volume for a given optical path length. For example, for Fig.21 , the constraints on the optical design now only require that the entrance mirrors 2101 do not overlap. Our current prototype exploits this idea, and this is Fig.18 and 19A -The reasoning behind the tapered pool design seen in C.

[0079] By utilizing Fig. 22 The elliptical beam pattern shown, the idea is extended to achieve even lower volumes. In this example, the entrance mirror 2201 has two holes 2204 as the entrance and exit (interchangeable). The surface on 2201 is spherical, but the profile of the mirror can be made rectangular or any other shape as long as it does not block the beam pattern. Note that in this embodiment, the beams are completely overlapping on the return mirror 2202. Not only that, but the beam reflection spots in the areas represented by 2203 and 2205 also overlap. This allows for a significant increase in the optical path length within a small volume, as shown by the envelope of the line shown by the laser path. Additionally, in vehicle exhaust applications, where inserting the probe into the flow would increase back pressure and adversely affect operation, the design allows for a radically low cross-sectional area blockage when the flow impacts a narrow surface. Note that this concept can also be used for any other non-elliptical beam pattern, such as Fig.23 In this figure, the mirrors are represented by 2301 and 2302, the spot pattern is represented by 2303, and the light trajectory is represented by 2304.

Claims

1. A laser absorption spectroscopy exhaust gas sensor, designed to operate at a temperature above 300°C, the laser absorption spectroscopy exhaust gas sensor comprising: an optical cell having porous walls with pores having an average diameter in the range of 10 nm to 10 microns; gold mirrors within the optical cell, positioned to support a multi-pass optical pathway within the optical cell, wherein each of the gold mirrors comprises a gold coating deposited on a substrate, a bonding layer sandwiched between the gold coating and the substrate, wherein the bonding layer comprises Ti or Ni; wherein the materials of the porous wall, the gold coating and the substrate are selected to operate at temperatures above 300°C; an active heating element adapted to heat the optical cell to a temperature above 300° C. to prevent condensation; A laser, suitable for generating a laser beam; an optical detector adapted to detect the returning laser beam; as well as A processor is isolated from the optical cell by a flexible conduit, the processor for controlling the laser and the active heating element and for analyzing signals from the optical detector to identify a gas in the optical cell.

2. The laser absorption spectroscopy exhaust gas sensor according to claim 1, characterized in that: The laser absorption spectroscopy exhaust gas sensor further includes an optical waveguide adapted to carry the laser beam from the laser to the optical cell and the return laser beam from the optical cell to the detector and to thermally isolate the laser from the optical cell.

3. The laser absorption spectroscopy exhaust gas sensor according to claim 1, characterized in that: The porous wall is composed of porous sintered metal powder, metal wire mesh or ceramics.

4. The laser absorption spectroscopy exhaust gas sensor according to claim 1, characterized in that: The porous wall has pores having an average diameter large enough to allow the gas desired to be sensed to pass through the pores and small enough to block undesirable contaminant particles from passing through the pores.

5. The laser absorption spectroscopy exhaust gas sensor according to claim 1, characterized in that: The substrate is fused silica.

6. The laser absorption spectroscopy exhaust gas sensor according to claim 1, characterized in that: The active heating element is integrated into the porous wall.

7. The laser absorption spectroscopy exhaust gas sensor according to claim 1, characterized in that: The active heating element is attached to the porous wall.

8. The laser absorption spectroscopy exhaust gas sensor according to claim 1, characterized in that: The wall of the optical cell has a non-porous portion.

9. The laser absorption spectroscopy exhaust gas sensor according to claim 1, characterized in that: The gold mirror consists of unprotected gold deposited on a substrate.

10. The laser absorption spectroscopy exhaust gas sensor according to claim 1, characterized in that: The gold mirror includes opposing spherical mirrors having different focal lengths.

11. The laser absorption spectroscopy exhaust gas sensor according to claim 1, characterized in that: The porous wall has a plurality of portions having different wall thicknesses, the plurality of portions being adapted to enhance gas flow into the optical cell.

12. The laser absorption spectroscopy exhaust gas sensor according to claim 1, characterized in that: The multi-pass optical path has a length greater than 50 cm and a physical length less than 5 cm.

13. The laser absorption spectroscopy exhaust gas sensor according to claim 1, characterized in that: The multi-pass optical path has substantially overlapping spot patterns on the mirror away from the entrance aperture and the exit aperture, the spot pattern having an elliptical shape on the mirror.

14. The laser absorption spectroscopy exhaust gas sensor according to claim 1, characterized in that: The substrate has a coefficient of thermal expansion that substantially matches that of gold.

15. A laser absorption spectroscopy exhaust sensor, comprising: Optical cell; a mirror within the optical cell positioned to support a multi-pass optical path within the optical cell, wherein the multi-pass optical path reflects from the mirror at spots on the mirror, wherein some of the spots distal to the inlet and the outlet significantly overlap each other; A laser, suitable for generating a laser beam; an optical detector adapted to detect the returning laser beam; as well as A processor is provided for controlling the laser and for analyzing the signal from the optical detector to identify a gas in the optical cell.

16. The laser absorption spectroscopy exhaust gas sensor according to claim 15, characterized in that: The light spot on the mirror forms a pattern having an elliptical shape.

17. The laser absorption spectroscopy exhaust gas sensor according to claim 15, characterized in that: The mirrors include opposing spherical mirrors having different focal lengths.

Images