Understanding particle counting technology
Understanding the principles behind the designs of optical particle counters can be invaluable in selecting the right system for a particular application
By Jim Babb, Adams Instruments
The original method for counting particles was to count them visually, through a microscope, a tedious, error-prone process that has mostly been replaced by optical particle counters. In absolute referencing visual particle counting must still be used. Understanding how optical particle counters work can be of great value in selecting the right instrument for a particular application.
Optical air particle counters detect particles by the Tyndall Effect, which is named after John Tyndall1 and usually applied to the light scattering of particles in colloid systems. The scatterings from dust in the air by a bright beam of light or fog are common manifestations of the Tyndall Effect.
Light is scattered when the refractive index changes. This means an air bubble in a liquid scatters light just as well as a solid particle in the same liquid. The way in which light is scattered by particles is described by what is called the Mie Theory.
The Lorenz-Mie-Debye theory was first published by Gustav Mie2,3 and describes how light is scattered in different directions. This changes with the refractive indices of the medium and the particle scattering the light as well as the particle’s size and wavelength of the light. It is outside the scope of this article to detail Mie theory; however, there exist various public domain applications4 that can be used to experiment with how light is scattered.
In the context of particle counters, the most important outcomes of Mie Theory and how it predicts light scattering relate to how the scattering varies with particle size. When the particle is much smaller than the wavelength of light, the light is scattered mostly in the forward direction (see Fig. 1a). When the particle is bigger than the wavelength of light, more light is scattered at right angles and backward (see Fig. 1b).
Light can be viewed as a wave that oscillates perpendicular to the direction of travel. This direction of oscillation is known as the polarization. The polarization of incident light is very important. In the previous examples, the light scattering is being measured in the same plane as the polarization of the incident light.
The scattering appears similar at 5 μm (see Fig. 2a); there are significant differences in the 0.3 μm particle’s scattering (see Fig. 2b) with polarization. The logarithmic scales hide any variations less than a factor of 10.
The amount of scattered light varies with the frequency: shorter wavelength = greater scattering. About 10 times more blue light is scattered than red light with all other things being equal. Most particle counters use a near infrared or red laser; until recently this was the most cost-effective option. Blue gas and semiconductor lasers are expensive; semiconductor lasers are also short-lived.
The air particle counter
The particle counter shown in Fig. 3 illustrates a typical sensor design; the airflow, laser, and collection optics are all at right angles to one another.
A vacuum attached to the sensor’s outlet draws air through the sensor. The laser light is scattered by particles in the air. This scattered light is collected by the optics and focused onto the photodetector, which converts the light into a voltage signal that is amplified and filtered. The signal is subsequently converted from an analog form into digital form for classification by a microprocessor. The microprocessor also interfaces the counter to a controlling data collection system.
Gas lasers were invented in 1960 and semiconductor lasers were developed in 1962. Although very expensive at first, when they became cost-effective gas lasers replaced white light in particle counters. The much less expensive semiconductor lasers later superseded these, for the most part, in the late 1980s.
Two types of lasers are used in particle counting: gas lasers such as helium-neon (HeNe) and argon-ion, and semiconductor lasers.5 Gas lasers are capable of producing intense monochromatic and sometimes even polarized light. The gas laser generates a collimated Gaussian beam and the semiconductor outputs a small divergent point source, typically with two different axes of divergence and all too frequently multiple modes. Due to the presence of multiple axes of divergence, the diode laser frequently has an elliptical output, which again presents challenges and some advantages. Different axes of divergence mean that one either concedes an elliptical output or devises a costly and complex series of optics to compensate. On the other hand, the elliptical beam lends itself well to certain applications by utilizing the long axis to attain better field coverage.
In summary, the output of a HeNe laser is “ready to use,” needing no additional optics. To generate a beam similar to a HeNe laser, the light from a semiconductor laser must be focused through lenses; this results in energy loss from the light source. However, the low cost, small size, low operating voltages, and modest power consumption make semiconductor lasers the preferred choice for particle counters.
In applications that require high sensitivity, HeNe lasers can be used in an open cavity mode to produce many watts of power (see Fig. 4).6 Because the sample is passed through the optical cavity, this type of laser fails at high particle concentrations due to the lasing action being quenched (failure to maintain the cavity “Q” factor).
The sample inlet to a particle counter plays a crucial role in the resolution of the particle counter. There are two styles of inlet: a flattened, wide (10 mm) but thin (0.1 mm height) version, and a round tube with an internal diameter of about 2 to 3 mm. With the flat style of inlet jet the laser beam is typically a narrow line in the same axis as the jet.
With the round style of inlet the laser beam is shaped to a line roughly at right angles to the axis of the inlet jet. The particles pass through a very narrow, intense sheet of laser light.
Each type of jet has its advantages and disadvantages. The air from a flat jet moving at a fairly uniform velocity and passing through the most intense and uniform part of the laser beam results in the best resolution.
However, the small cross-section means a higher vacuum is required than the round jet, which increases power consumption (important especially with battery-powered units). Flat jets are more complex and costly to manufacture, and the alignment with the laser is problematic.
The simpler round jet, because of its larger cross-section, requires a lower vacuum for the same flow rate, so less power is consumed when air is drawn in. Slower airflow also means more light is scattered per particle than with a flat jet. The disadvantage to the round jet is the reduced uniformity of airflow and the variations in laser power across the beam; the beam is stretched out, resulting in poorer resolution.
Particles scatter light in all directions, mostly in the forward direction. As the particle becomes bigger, more light is scattered backward and at right angles. Collection optics gather and focus the light onto a detector, avoiding laser interference.
Collection optics also remove unwanted light by attempting to gather only the rays that contain desired signal. Light from stray reflections causes noise, usually seen as a baseline offset, and reduces the instrument sensitivity.
Reflectors: Concave mirrors can be used to collect light and focus it onto a detector. A type of concave mirror known as a lamp reflector can reflect light emitted from its focal point right back to the focal point. These are the most commonly used type of collection optics because they allow small, compact sensors to be made at low cost.
Lenses: The lenses used in particle counters are frequently aspheres used in pairs. They effectively move an image (the scattered light) from one focal point to another (the photodetector). In many sensors a reflector also is used to collect the light from the other side of the lenses.
By the careful use of masking techniques such as limiting apertures or field stops, stray light can be further reduced. The use of lenses to transfer light from one plane to another and the stray light reduction techniques are not unlike those used in typical photography, but keep in mind that particle counters employ monochromatic radiation and therefore do not have to worry about additional chromatic aberration correction (multiple wavelengths of light focus at different points when refracted).
Mangin mirrors: A Mangin mirror consists of a meniscus negative lens with a mirrored convex second surface. These were commonly found in acetylene-powered lamps. Now they are used in optical systems such as telescopes.
Mangin mirrors are used in particle counters as pairs like aspheres. The mirrors are lighter but wider than lenses. As with aspheres, the function is to move an image from the focal point of one mirror to the focal point of the other.
Non-imaging particle counters: A non-imaging particle counter does not use any collection optics. It is a photodetector placed close to the sample inlet and laser that collects scattered light. Small sensors (e.g., in handheld units)-although often including optics-have an element of non-imaging in their operation.
The photodetector converts the incident light as photons into electrical pulses by creating a charge for each received photon. As the amount of scattered light increases with the particle’s size and the scattered photons arrive at the same time, a current pulse proportional to the particle’s size is generated.
Photodiodes: A photodiode is a p-n junction. When a photon of sufficient energy strikes the diode, it creates a mobile electron and a positively charged electron hole. These charges give rise to the photocurrent, which is then is amplified, filtered, and classified.
Avalanche photodiodes: An avalanche photodiode7 is a semiconductor version of a photomultiplier. One photon can trigger an avalanche of electrons in the device; it is possible to detect and count single photons. However, these devices run on high voltages (hundreds of volts), cost many times the price of a photodiode, and require relatively complex circuitry to work at the speeds required for particle counting. Not surprisingly, these devices are used only in high-sensitivity equipment.
The signal processing electronics amplify and filter the signal from the photodetector.
For example, the (exaggerated) signal in Fig. 5a could have come from a particle counter. There are four peaks from particles. The variations in the baseline could be due to acoustic pickup (e.g., from the pump), power supply, or what amounts to whistling as the air is drawn at high speed through the inlet jet. This is removed by high-pass filtering that removes signals with a frequency much lower than those from the particles.
This leaves high-frequency interference that could come from the electronics, for example (see Fig. 5b). Low-pass filtering removes signals with a frequency much higher than those from the particles.
After the signal has been cleaned, it consists of a series of pulses, the height of which is related to the size of the particle (see Fig. 5c). These signals are now classified, converting them from analog to digital form by using a pulse height analyzer. When converted to digital form, the classified pulses can be counted and finally reported to some controlling system.
Jim Babb is the director of optical engineering at Adams Instruments. Over the past 24 years, he has been involved in the development of highly complex laser electro-optical systems and defining metrology standards for the FDA, defense contractors, and aerospace manufacturers. He can be reached at firstname.lastname@example.org.
- See http://en.wikipedia.org/wiki/John_Tyndall.
- Kerker, M., The Scattering of Light and Other Electromagnetic Radiation, Academic Press, New York, pp. 54-59, 1969.
- Mie, Gustav, “Beiträge zur Optik trüber Medien, speziell kolloidaler Metallösungen,” Annalen der Physik, Vierte Folge, Band 25, No. 3, pp. 377-445, 1908.
- See http://atol.ucsd.edu/scatlib/.
- See http://en.wikipedia.org/wiki/Laser_diode.
- Schuster, B.G., and R. Knollenberg, “Detection and Sizing of Small Particles in an Open Cavity Gas Laser,” Appl. Opt. 11, p. 1515, 1972.
- Dautet, H., et al, “Photon-counting Techniques with Silicon Avalanche Photodiode,” Appl. Opt. 32 (21), pp. 3894-3900, 1993.
Some of the graphics used in this article were obtained from Wikipedia (www.wikipedia.org) and are in the public domain.