GB2264556A - Diffraction analysis of particle size, shape and orientation - Google Patents

Abstract

In apparatus for the analysis of individual aerosol particle characteristics, a nozzle directs a particle stream past a polarised laser beam, a video camera comprising a 2-D sensor array, e.g. a CCD, images light scattered from the stream of particles, and a data processor analyses captured video frames to deduce particle shape, orientation, and size from the diffraction patterns. Video frames are captured in response to a trigger from an auxiliary detector. Image intensifiers and expert system particle identification is discussed. <IMAGE>

Description

Analysis of particle characteristics The ability to measure particles by shape and size is of importance to many groups of people. The food and chemical industries are concerned from a quality control point of view; biologists are interested in characterising cells and monitoring changes in and differences between cells; environmental scientists are concerned with airborne particles and their effect on air quality and health. This list is by no means exhaustive, it is merely intended to illustrate the driving force behind the attempts to develop accurate and reliable measurement instrumentation, and to theoretically understand the nature of the problem.

There are currently two main optical scattering methods in use in commercially available particle measurement systems. The first method attempts to size particles by measuring their static or dynamic behaviour in fluid. These systems generally measure deposition rate, acceleration in a jet stream or Brownian motion. The second method attempts to size particles by measuring the light scattered from an illuminated particle or ensemble of particles either at a few specific angles or over a large solid angular range. Apart from image analysis systems, none of the commercial instruments is capable of characterising particles by shape, non-spherical particles being sized by assigning an equivalent spherical diameter, although this diameter depends on the measurement method used.What is worse, is that some instruments are known to become inaccurate when tested with non-spherical particles of regular shape, so measurements taken with particles of arbitrary shape have to be treated with some caution.

Instruments which attempt a shape classification are based on image analysis, which requires taking an image of a small number of particles and performing complex image processing. The particle sample has to be prepared beforehand so that it is in a form suitable for image processing, i.e. it has to be processed so that individual particles can be seen with minimal overlapping. Thus there is a considerable time delay before the results are available. The method also requires fast computers in order to do the analysis reasonably quickly. Some of the other instruments also suffer a time delay before measurements are available, and whether this is important depends on the application. It is not necessarily important for batch testing powders for example, but it is of potential importance when monitoring a working environment for asbestos fibres or micro-organisms.

Several commercial laser based instruments are available which will size particles, as disclosed in "Particle Size Analysers Product Roundup". Powder and Bulk Engineering, Feb 1991; pp 42, and other research instruments have been built to investigate various aspects of particle sizing. For example, an instrument has been developed to size particles using the oscillation in intensity of the scattered light, as disclosed in "Drop Sizing by Laser Light Scattering Exploiting Intensity Angular Oscillation in the Mie Regime." by Ragucci, R., Cavaliere, A. and Massoli, P. Particle and Particle Systems Characterisation, Vol 7, 1990; pp 221. Most of the instruments analyze an ensemble of particles and, as stated previously, they assume a spherical particle or particles, and do not give any indication of non-sphericity.

Research reported in "Light Scattering Instrument to Discriminate and Size Fibres Part 2: Experimental System". Particle and Particle Systems Characterisation, Vol 6, 1989; pp 144, has been reported using an instrument designed to discriminate and size fibrous material. In this research, particles are passed through a laser beam in single file using a laminar airflow system similar to the design desribed below. The forward scattered light is collected by a lens and passed through a polarizing beamsplitter. The intensity of the light in two orthogonal polarizations is then recorded using photo-multiplier tubes. Results show that near spherical particles can be discriminated from fibrous particles by taking the ratio of the polarized intensities, provided that the particle diameter is above 1.5 microns approximately.

An instrument which has been developed to size particles uses the laser Doppler velocimetry technique of "Strengths and limitations of the phase Doppler technique for simultaneous measurements of particle velocity and size." by Livesley, D.M. Proceedings-SPIE International Society for Optical Engineering Vol 952, 1988; pp-454, and this has also shown a capability of discriminating near spherical particles from fibrous ones. This technique is based on refraction of rays by the particle, so it is limited to particles larger than 5 microns. The instrument uses two coherent laser beams which interfere with each other, creating a series of fringes in the scattering volume. The spacing of the fringes depends on the wavelength of the lasers and the angle between the beams.Three photo-multiplier tubes are used at different angles of forward scatter which together give an indication of particle speed, size, and non-sphericity.

The speed is obtained from the time it takes the particle to traverse from one fringe to the next. The size is obtained from the phase difference in the signals from two detectors, which is a function of speed and particle surface convexity, the rate of sweep of the refracted ray as the particle traverses the fringe being larger for a smaller particle. The third detector allows a second phase difference to be measured, and a different in the two measured phases is seen when the particles are non-spherical.

We have disclosed in "An instrument for the classification of airborne particles on the basis of size, shape, and count frequency." Atmospheric Environment, Vol 25A No. 3/4, 1991; pp 645. by Kaye, P.H., Eyles, N.A., Ludlow, I.K., and Clark, J.M., an airborne particle classifier (APC) which has some capability of determining particle shape as well as size: it is shown in Figure 1 of the accompanying drawings. In Figure 1, the components are labelled thus: 1. HeNe Laser 2. Main-Chamber 3. Rear-Chamber 4. Photomultiplier Housing 5. Beam Expander 6. Iris Diaphragm 7. Cylindrical Lens 8. Scattering Volume 9. Filter for sheath-air 10. Ellipsoidal reflectors In this instrument, particles are passed through a laser beam in single file, and the scattered light is collected by an ellipsoidal reflector surrounding the particle. The scattered light then passes through a lens before impinging on three photo-multiplier tubes arranged in a triangular configuration about the principal optical axis of the system. The outputs of the photo-multiplier tubes are digitized and stored using dedicated electronics and analysed using a computer. The system is capable of collecting information on a maximum of 10,000 particles per second, and is thus capable of quasi-real time operation.

However, the shape information is severely limited because of the small number of detectors, and it is unlikely that it could be used to differentiate unambiguously between different types of non-spherical particle, e.g. fibres and platelets. There is also uncertainty in the trajectory and orientation of particles as they pass through the beam, and it is difficult to determine and allow for the effect of these on the scattering with only three detectors.

Thus no real-time method of shape analysis is yet available, and little investigative work has been done on non-spherical particles.

Accordingly, the present invention is intended to overcome such limitations for the purposes both of classifying known particles, and analysing unknown or test particles in a practical instrument.

The invention provides apparatus for the analysis of individual particle characteristics (such as shape and size) from an aerosol or other suspension of particles comprising a monochromatic light source, a scattering chamber with means for directing a stream of the particles through a collimated beam of light from the light source, a detector including a two-dimensional array of a multitude of sensors arranged to image light scattered from the stream of particles, the detector capable of resolving diffraction or interference maxima and minima of light scattered from particles of a size in the range required, and a data processor for capturing successive sets of image data from the detector and representing the corresponding images.

The invention also provides a method of analysing characteristics of individual particles, such as shape and size and orientation, from an aerosol or other suspension of the particles, comprising passing a stream- of the particles across a collimated beam of monochromatic light and imaging diffraction or interference maxima and minima in the scattered light using a detector, capturing the image data in a data processor and processing those data in accordance with predetermined criteria.

Preferred embodiments of the invention are intended to be capable of classifying particles into one of five broad shape classifications: spheres, droplets, fibres, platelets and "chunks" (i.e. particles of comparable size in all three dimensions), and also of differentiating between particles of differing aspect ratio.

In this specification, the term "scattering profile" is intended to mean the three-dimensional scattered light intensity distribution about the particle. The scattering profile is unique for particles of given shape, orientation, and dielectric structure for a given wavelength of illumination.

In order that the invention may be better understood, an example will now be described in detail with reference to the accompanying drawings, in which: Figure la, to which reference has already been made, is an elevation of a conventional APC scattering chamber; Figure lb is a corresponding plan view; Figure 2 is a diagram of a system embodying the present invention:: Figure 3 is a sectional view showing the scattering chamber assembly of Figure 2; Figures 4 to 7 are ray trace diagrams illustrating the passage of light in the scattering chamber; Figure 8 is a sectional view through the inlet assembly, between the particle collection chamber and the scattering chamber of Figure 2; Figure 9 shows the collected image of light scattered from a 1 micron polystyrene sphere between 300 and 140 ; Figure 10 shows the collected image of light scattered from a water droplet between 300 and 1400; and Figure 11 shows the collected image of light scattered from an arbitrary airborne particle between 5 and 300.

Due to the considerable promise shown with the APC described with reference to Fig. 1, the new instrument shown in Figs. 2, 3 and 8 retains the same ellipsoidal reflector light collection system and the same concentric tube type of particle delivery system. The light collection system has the capability of collecting light scattered into 82% of the total sphere of scattering around the particle. The concentric tube delivery system can be arranged to constrain particles to a given scattering volume diameter, and has shown itself to be capable of some particle alignment.

The main change required to the APC of Figure 1 was to increase the spatial resolution of the instrument. In accordance with the invention, the solution was to use a charge coupled device (ccd) video camera coupled with an image intensifier. This arrangement gives four advantages: 1. The resolution is increased by replacing three detectors with the 110880 (385 by 288) elements of the ccd array.

2. The light input to charge output conversion efficiency is virtually constant across the array because it is manufactured in one piece of silicon.

3. The light gain of the image intensifier is also virtually constant over its aperture due to its construction.

4. Use of a video camera means that images can be captured on a standard computer frame grabber board, and therefore can be processed and stored faster than with a still camera using photographic film.

The image intensifier is required in this example of the invention because of the small number of photons scattered from a particle during its transit through the laser beam. The time of transit of a particle through the beam is approximately (2-5) microseconds depending on flow rate, during which time typically several thousand photons are scattered depending on particle size, when using a focused 10mW helium-neon laser as in the APC. The number of photons scattered depends approximately on the fourth power of the particle size in the size range of interest (approximately 1-10 microns), and hence varies considerably. The camera has an asynchronous trigger facility which allows it to be used in a similar manner to a still camera. This is necessary so that the scattered light can be captured during the time that a particle is in the beam.

The maximum number of images that the camera can output is 25 per second, i.e. it can record data on 25 particles per second, as opposed to the data rate of 10,000 particles per second of the original instrument.

This is not a disadvantage however, since the instrument in this example is used for basic research into scattering profiles from particles of different shape. The high spatial resolution offered by the camera allows the determination of an optimal detector configuration for use in real time instruments with more rapid particle handling rates, also embodying the invention. Such an optimised detector configuration may well have fewer elements than a camera CCD array so as to allow a higher particle analysis rate. The configuration need not be symmetrical, and may be optimised to suit analysis of a specific particle shape. Examples would include a custom multi-element photodiode array arranged geometrically for optimal shape characterisation.

In the particle size range of particular interest (1-10 microns), a significant proportion of the total scattered light is scattered in the angular range up to 300. This light passes through the hole in the ellipsoidal reflector, and is thus not collected by the camera. Light scattered in this angular range may be of importance in determining particle shape and hence it was decided that the instrument should be designed so that this angular range could be investigated if necessary.

The light passing through the hole is normally collected by a photo-multiplier tube which generates a pulse to trigger the camera, and it was therefore decided to design the scattering chamber to allow the camera and photo-multiplier tube housing to be interchanged. This arrangement allows the camera to record scattering in either the range 5 -30 or the range 300-1400.

Previous results recorded from fibres have shown that there could be differences in orientation between particles as they traverse the gap between the inlet and outlet tubes. To investigate this more thoroughly it was required that the distance between the inlet and outlet tubes be adjustable, and that the distance between the ends of the tubes and the laser beam be made adjustable.

It was also decided to make the depth of the particle delivery tube inside the concentric clean air tube adjustable so that the aerodynamic focusing effects of this system could be investigated.

In the original instrument the clean air filter was an integral part of the instrument, which meant that only total airflow could be monitored. In the new example of Fig. 2 it was a requirement that the clean air flow rate could be monitored separately as well as the total, so that the ratio between particle laden air and clean air could be monitored and adjusted.

Since the camera outputs a standard CCIR video signal, a standard frame grabber board is reqtiired to capture the frames, and the camera. trigger interface has a purpose-built board. The system therefore has a micro-computer which enables the grabbing and storing of frames to be controlled from the keyboard. A computer performs any post processing of the images which may be necessary.

The system configuration is shown in Figure 2, and comprises the scattering chamber, laser, pump, camera, photo-multiplier tube (12, Figure 3), and computer with frame grabber and trigger controller boards.

The aerosol sample is drawn through the scattering chamber by a vane-type pump which can be adjusted to give different flow rates. When a particle traverses the laser beam, the scattered light is picked up by the photo-multiplier tube, the output of which is used to trigger the camera via the camera trigger controller board. The output of the camera is then captured on the frame grabber board and output to a monitor and stored to disc if required. Both boards are under purpose-designed software control, enabling easy operation of the system from the computer keyboard.

When storing images to hard disc the image capture rate is approximately two images per second due to speed limitations of the disc.

Figure 3 is an assembly drawing of the scattering chamber in side view, in which the components are labelled thus: 1. Laser beam 2. Laser beam shroud 3. Hole for laser assembly 4. Particle inlet assembly 5. Particle outlet assembly 6. Ellipsoidal reflector 7. Optical window 8. Iris 9. Back scatter lens assembly 10. Forward scatter lens assembly 11. Camera 12. Photo-multiplier tube 13. Scattered light rays The laser beam 1 enters through a hole 3 in the side of the chamber and is turned through 900 by a silvered prism which is glued to an optical window 7. This method of mounting was chosen so that no shadows are created by prism mounting post. The beam 1 then passes through a shroud 2 and through the focus of an ellipsoidal reflector 6 before being deflected through 900 by a second silvered prism which is glued to a lens 10.The shroud is necessary to reduce background off-axis scattering caused by imperfections on the mirror surfaces. The beam is then absorbed by the matt black surfaces behind the ellipsoidal reflector.

The particle inlet 4 and outlet 5 tubes are arranged so their axes coincide with the primary focus of the ellipsoidal reflector 6. Particle laden air is drawn through the chamber at (2.5-6)1/m creating a columnar, laminated airflow with a diameter of lmm across the gap between the tubes. Particles thus pass in single file through the laser beam within 0.5 mm of the focus of the ellipsoidal reflector.

Light 13 scattered between 280 and 1410 from a particle is collected by the ellipsoidal reflector. This is mounted in a cylindrical holder which can be moved to position it correctly. This is necessary because although the optical and mechanical dimensions of the reflector are specified, no relationship between them is given. The reflected light passes through the optical window 7 and an iris 8 positioned at the secondary focus of the reflector. The iris is necessary to reduce the amount of background scatter which would otherwise reach the camera 11. The effect of the optical window is to increase the distance between the two foci of the reflector by approximately 6 mm. The light then passes through a pair of plano-convex lenses 9 to the faceplate of the camera.

Light scattered between 5 and 280 from a particle passes through a pair of plano-convex lenses behind the ellipsoidal reflector and then through an iris to the faceplate of the photo-multiplier tube 12. The camera mounting and the photo-multiplier tube mounting can be interchanged to record images at both these angular ranges.

The optical system was designed using a dedicated ray tracing package, so that the effects of particles passing through the scattering volume at different positions could be determined. This was necessary because the ellipsoidal reflector causes the images to be a non-linear representation of the scattering angle, and this in turn is a function of particle position. Figure 4 shows a result of a ray trace from the package with the optical configuration used in the instrument. The rays correspond to scattering angles 30 to 1400 in 5 steps.

The ray trace shows the complete system with rays originating from a particle at the focus of the ellipsoidal reflector. The package allows the user to zoom in on any portion of the trace, so a detailed examination is possible. Figures 5, 6 and 7 show details of ray traces at the output of the lens system with particles positioned at the focus of the ellipsoidal reflector, and 0.5 mm either side of the focus. In Figure 6, the particle is 0.5 mm towards the secondary focus; in Figure 7, it is 0.5.mm away from the secondary focus. The figures clearly show the sensitivity of the instrument to changes in the position of the particle, but by positioning the camera on the y axis in the figures the effect of the sensitivity on the image is minimised. The rays correspond to scattering angles 300 to 1400 in 50 steps.

The laser assembly is a Lasermax model LAS-200-670-10 diode laser module with integral power supply and collimating optics. The laser has a power output of 10 mW at a wavelength of 670 nm. The output is plane polarized and operates in the TEM,, mode, with a cross section of 4 mm by 1 mm. The module is mounted in the housing by set screws which enable it to be aligned with the chamber. The beam passes through a quarter wave plate to render it circularly polarized, through an iris diaphragm and then through a cylindrical lers before entering the main chamber.The beam at the focus of the ellipsoidal reflector is thus approximately elliptical with dimensions 3 mm wide and 120 microns ce~?. The laser housing can be moved inside the mounting te allow for cylindrical lens focal length tolerances.

The particle inlet assembly is shown in figure 8.

The system utilises a concentric tube arrangement which creates a laminar airflow in the scattering volume, the particle laden air being surrounded by a sheath or clean air. Clean air from a 0.2 micron filter enters a small chamber via two pipes before being drawn into the scattering chamber through the outer tube. This arrangement ensures low air velocity in the aii camber to prevent the formation of vortices. T'e chamber sub-assembly can be moved within its housing by a threaded adjuster to change the depth of the tubes in the scattering chamber.

Particle laden air is drawn into the scattering chamber through the inner tube 2 which is fixed into a second assembly mounted on top of the clean - number sub-assembly. This assembly controls the position of the inner tube in relation to the outer tube 1, tre tube holder again being moved within its housing by means of a threaded adjuster. In Figure 8, the components are labelled thus: 1. Sheath air tube 2. Particle laden air tube 3. Sheath air chamber 4. Sheath air inlet 5. Depth adjusting nut for inlet tubes 6. Locking rings 7. Particle air tube carrier 8. Depth adjusting nut for particle air tube 9.Particle laden air inlet The particle outlet assembly is very similar to the inlet assembly, enabling adjustment of the depth of the tube in the scattering chamber. The tube is fixed in its holder by two nylon bushes, thus electrically isolating the tube and allowing a voltage to be applied to it if desired. The effects of a longitudinal electric field on non-spherical particle alignment can thus be investigated The camera interfaces to the computer using two expansion boards. The first board is a commercial frame grabber board with a 256 by 256 pixel array which captures the camera output.

Commercial software is available with this board to enable any post processing of the images to be performed.

In particular, this software may include an expert system for classifying spatial distributions of the captured video frame diffraction maxima corresponding to known particles of different shapes and sizes, and for then identifying those characteristics in test particles.

The analysis may include recording features of the spatial arrangement of the image maxima for each of several particles for whom the relevant characteristics are known, and then using those features to set the predetermined criteria for analysis of test particles.

The second board is newly designed to control the camera triggers and frame grabber and has four main functions as follows: 1. Trigger generation. The board uses the output of the photo-multiplier tube to generate two trigger pulses, one for the image intensifier and the other for the video camera. The image intensifier is turned on for the duration of time that the particle is in the scattering volume and thus acts as an electronic shutter. The camera trigger pulse is of 3 ms duration and starts after the trigger is removed from the image intensifier. This timing arrangement was found to.improve the quality of the images by reducing ccd overspill between elements (smearing). The persistence of the image intensifier phosphor ensures that no data are lost.

The trigger pulses can be disabled from the keyboard.

2. Noise floor setting. The level below which the camera will not be triggered can be set from the keyboard. This prevents the camera being falsely triggered by noise from the photo-multiplier tube.

3. Time of flight checking. The board contains a counter which records the length of time that a particle is in the laser beam to a resolution of 125 ns. This can be read from the keyboard or by programs and is used to discriminate between genuine single particles and either particles floating into the beam outside the scattering volume, or two or more particles following each other through the beam. It is not possible to detect if two or more particles pass through the beam alongside each other, unless this is evident from the image generated by the camera.

4. Interrupt generation. The camera has a digital output which signals when it is about to output an image. This is used by the board to generate interrupts at the start and end of the video frame.

A software interrupt routine then sets up the frame grabber board to capture the frame.

Figures 9, 10 and 11 show some images taken with the present system. The laser has a linearly polarized output so the scattering profiles are different in different scattering planes. This is because the scattering profile in any plane is dependent on the parallel and perpendicular components of the polarization of the incident beam, which are different for each plane.

The images from polystyrene spheres, water droplets and laboratory air have been captured. The images exhibit a grainy nature, especially at larger scattering angles, which is due to the small number of photons being scattered during the time the particle is in the beam.

Each grain on the image corresponds to an individual photon incident on the faceplate of the image intensifier.

The images show spatial scattering profiles from approximately 300 around the central shadow region to approximately 1410 around the edge of the pattern limit.

The central shadow region is caused by the hole in the ellipsoidal reflector and its size varies depending on particle position in the scattering volume as predicted by the ray tracing package. The uncertainty in the scattering angle limits is because the system has not yet been calibrated to take into account changes in particle position, and it is known that the instrument is very sensitive to changes in particle position, particularly at small scattering angles.

The other shadow regions are caused by the inlet and outlet tubes. The almost circular portions are caused by the ends of tubes intercepting light before it reaches the ellipsoidal reflector, the narrow portions are caused by the sides of the tubes intercepting light after it has been reflected by the reflector. The 900 scattering angle is a circle whose circumference intercepts the centres of the circular shadow regions.

Figure 9 shows a result from a 1 micron diameter polystyrene latex sphere. Concentric rings of scattering maxima and minima can be seen, which is consistent with the theory of scattering from spheres.

There is a good qualitative agreement between the image and theoretical predictions. In particular, it can be seen that the minimum which occurs at an angle of approximately 70" in the horizontal plane disappears. in the vertical plane, which agrees with theoretical predictions. The scatter angles at which the maxima and minima occur are also in good agreement.

Simulations with 2.95 micron and 4.3 micron diameter spheres have also been carried out with similar correlation, although the high periodicity of the fringes and the non-linear nature of the experimental data make quantitive comparison at this stage more complex.

Figure 10 shows a typical result from a water droplet generated using a water spray, with scattering between 300 and 1400. Since the spray generates a wide droplet size distribution, the diameters of the individual droplets creating the scattering profiles are not known.

Simulations of scattering profiles from 2 micron and 2.5 micron diameter spherical water droplets have been carried out and these show that relatively more light is scattered in the 900 to 1490 angular range than is the case with polystyrene spheres. This can also be observed in the experimental data.

With the camera positioned to capture small scattering angle data from 50 to 300, the laser was fitted with a 22% transmission broadband neutral density filter on its output to reduce the beam intensity, because -without- this, the amount of scattered light is such that the ccd array is saturated, and the image is normally completely white for droplets of the size range generated.

Figure 11 shows an image taken when the instrument is operated with the inlet open to the atmosphere in the laboratory.

To summarise, the instrument possesses the capability of resolving diffraction maxima and minima with particles in the approximate size range (1-10) microns.

The lower limit is governed by the number of scattered photons, which varies approximately in proportion to the fourth power of the diameter of the particle for particles of this size, and therefore changes by several orders of magnitude in the size range of interest. The lower limit could be reduced by using a more powerful laser but this would cause camera saturation problems with larger particles. The upper limit is caused by the number and closeness of the scattering maxima and minima and could be increased by using a camera with a larger ccd array and redesigned instrument optics.

The nature of the ellipsoidal reflector causes scattering angles below approximately 700 to be compressed into a relatively small area on the ccd array, and therefore detailed examination of this region could be facilitated by redesigning the lens system to expand this region, at the expense of losing scattering information at larger angles. The instrument would readily accommodate such changes.

The instrument is sensitive to changes in'particle position in the scattering volume, especially at small scattering angles. While this is a disadvantage when calculating the angles where diffraction maxima and minima occur, it is possible to take this into account by observing the size of the central shadow region. There will, however, always be an uncertainty in the calculations, especially at small angles.

The apparatus can also be exploited to determine the orientation of particles as they traverse the airflow between the inlet and outlet tubes. The two tubes can be moved in relation to the laser beam, which effectively enables the scattering volume to be moved to different positions of the airflow across the gap. The distance between the tubes can also be changed so that airflows of different lengths can be investigated, and an investigation of the aligning effects of an electrostatic field can be carried out by putting a voltage on the outlet tube, which is electrically isolated from the rest of the instrument.

Although it is not possible to trace one particle across the complete airflow, it is possible to determine whether there is any preferred orientation of particles at any particular position in the airflow, and if so, whether this orientation changes as the particle moves with the flow.

From an investigation of the aerodynamics it should be possible to design a particle delivery system which ensures an almost constant orientation. This is seen to be a requirement in a practical instrument capable of shape discrimination since the number of detectors could be much less than on the research instrument of Figure 2, and changes in orientation would almost certainly produce an incorrect shape result.

The apparatus could also be used to investigate the deformation of liquid droplets as they traverse the airflow. If any deformation can be detected, it may be possible to differentiate between solid spheres and droplets.

Droplets of known size would be generated using an aerosol generator, and experiments performed with liquids of different viscosities recording scattering profiles as droplets traverse the gap between inlet and outlet tubes.

One could then determine if there is any elongation or relaxation of droplets due to shear forces in the airflow across-the gap, which would cause a non-spherical scattering profile. From the scattering profiles already obtained from water droplets, it is expected that any non-spherical scattering profile due to deformation will be small, and detection will almost certainly require computer analysis. Appropriate algorithms would detect deformations once the scattering profiles have been obtained.

Theoretical modelling of the aerodynamics of the particle delivery tube system can also be used, and compared with the experimental results. It should be possible then to determine an optimum delivery system so that droplets can be discriminated from solid spheres and non-spherical particles.

Although the apparatus has been used so far with aerosols in air or other gas, the apparatus could be used to detect characteristics of particles supported in a fluid or liquid medium, such as a colloidal suspension, with minor modifications to the chamber and optics.

Claims (16)

1. Apparatus for the analysis of individual particle characteristics (such as shape and size) from an aerosol or other suspension of particles comprising a monochromatic light source, a scattering chamber with means for directing a stream of the particles through a collimated beam of light from the light source, a detector including a two-dimensional array of a multitude of sensors arranged to image light scattered from the stream of particles, the detector capable of resolving diffraction or interference maxima and minima of light scattered from particles of a size in the range required, and a data processor for capturing successive sets of image data from the detector and representing the corresponding images..

2. Apparatus according to Claim 1, in which the detector is a video camera.

3. Apparatus according to Claim 1 or 2, in which the data processor is programmed to perform a spatial analysis of the diffraction maxima and minima in the images, in accordance with predetermined criteria, to determine the particle characteristics.

4. Apparatus according to Claim 1, 2 or 3, in which the camera is capable of resolving diffraction maxima and minima from particles in the diameter range 1 to 10 microns.

5. Apparatus according to Claim 1, 2, 3 or 4, comprising an image intensifier in-the light path between the particle stream and the camera.

6. Apparatus according to Claim 1, 2, 3, 4 or 5, in which the light source is a continuous wave laser.

7. Apparatus according to any preceding Claim, comprising means responsive to the presence of a particle in the light path to produce a presence signal, and in which the data processor responds to the presence signal to trigger the detector to capture a set of image data.

8. Apparatus according to Claim 7 as dependent on Claim 5, in which the data processor also responds to the presence signal to turn on the image intensifier for the duration of the particle in the light path so that it acts as an electronic shutter.

9. Apparatus according to Claim 8, in which the data processor triggers the detector only after the image intensifier has been turned on and off, the persistence of the image in the image intensifier then providing the necessary image for the detector.

10. Apparatus according to Claim 7, 8 or 9, in which the data processor records the duration of the presence signal and is programmed for analysis of that duration including discrimination between single particles from the stream, a series of two proximate particles from the stream, and spurious particles.

11. Apparatus according to any preceding Claim, in which the data processor comprises an expert system for classifying spatial distributions of the captured image diffraction maxima and minima corresponding to known particles of different shapes and sizes, and for then identifying those characteristics in test particles.

12. Apparatus according to any preceding Claim, in which the detector has a ccd array of at least 256 x 256 pixels.

13. A method of analysing characteristics of individual particles, such as shape and size and orientation, from an aerosol or other suspension of the particles, comprising passing a stream of the particles across a collimated beam of monochromatic light and imaging diffraction or interference maxima and minima in the scattered light using a detector, capturing the image data in a data processor and processing those data in accordance with predetermined criteria.

14. A method according to Claim 13, including recording features of the spatial arrangement of the image maxima and minima for each of several particles for whom the said characteristics are known, and then using those features to set the predetermined criteria for analysis of test particles

15. Apparatus for analysing particles, substantially as described herein with reference to the accompanying drawings.

16. A method of analysing particles, substantially as described herein with reference to the accompanying drawings.