For Technophiles

Using Photometry to Measure Mean Fibre Diameter (1950 - 1970)

Principle

Photometry is the analytical use of light (luminous) intensity to measure the physical and chemical properties of solids, liquids and gases, and mixtures or solutions thereof. Wavelengths in the infrared, visible and ultra-violet portions of the electromagnetic radiation spectrum are commonly used in photometric measurements. Photometry is probably the most extensively used of all analytical technologies.

In principle the application of photometry to the measurement of wool fineness is very simple. Photometers consist of a source of light of constant radiance, a sample cell and a photo-detector. In the specific case of wool the measurement of fineness is based on the principle of light scattering and presumes that the wool fibre is opaque. The impact of the light beam on the photo-detector in the absence of any interference will generate a detectable electrical signal. If a fibre at right angles to the beam intersects the beam the fibre will project a shadow onto the photo-detector, due to the light incident on the fibre being scattered. The shadow of the fibre will reduce the signal from the photo-detector by an amount that is proportional to the projected area of the fibre. If the length of the fibre intersecting the beam is constant then the output from the photo-detector will be proportional to the transverse dimension of the fibre¹.

However, this is a very simplistic description of the physics of photometry as applied to wool fibre diameter measurement. The specific details of the physics in particular instruments are dependent upon the instrument design.

When measuring wool fibre diameter the critical step required for photometric instruments is a technique for aligning the fibres so that they are always at right angles or near right angles to the beam of light, and of ensuring that the orientations of the individual fibres across the beam are similar.

There is an extensive literature describing aspects of photometric techniques for the measurement of the diameter distribution characteristics of wool i.e.

• Mean Fibre Diameter (MFD);
• Standard Deviation of Diameter (SD); and
• Coefficient of Variation of Diameter (CVD).

The topic is simply too big to cover in one article. Because of this we will be considering the use of Photometry in several parts, each part focusing on key developments within specific time periods i.e.

• The First Efforts (1950-1970);
• Leaping Forward with Laser Optics (1970-1983);
• Resolving Problems (1983-1989); and
• The Sirolan Laserscan (1989-today).

The first two periods are covered in this issue.

The First Efforts (1950-1970)

The earliest development of a photometric instrument began in the United States. The issue of aligning the fibres appropriately was resolved in this instrument by incorporating a device for aligning fibres on a microscope slide called the Fibroalineator.

Developed in the early 1950’s the Fibroalineator was an extension of a technique first reported by Larose (1947), for aligning fibres on a microscope slide by applying an electric field across the slide.

In 1957 the Sheep and Fur Animal Branch, Animal Husbandry Research Division, ARS, USDA in Beltsville Maryland commenced work on a prototype photometric instrument, which incorporated the Fibroalineator. The instrument became known as the Electronic Fibre Fineness Indicator (EFFI).

The instrument was an optical, mechanical and electronic system designed to automatically scan a microscope slide on which cut wool fibres had been mounted. The prepared slide was placed on a movable microscope stage equipped with high voltage electrodes, which by means of electrostatic forces caused the fibres to align parallel to the electrostatic field – the principle of the Fibroalineator. A beam of light from an incandescent lamp passed through a condensing lens and then a polarising filter. The polarised light then passed through the wool fibres on the slide on which the fibres were aligned by the electrostatic field. The wool fibres being bifringent rotate the light that passes through them. A microscope was located below the slide and magnified the image of the fibre before the light beam passed through a second polarisingfilter located below the microscope. This filter was mounted at right angles to the first, and blocked the unrotated light, allowing the rotated light to pass. Thus the microscope projected an image of the fibres against a black background onto a slit.

A synchronous motor was programmed to move the slide carrier from right to left and to move the slide carrier forward 1 mm at the end of each crossing. The slide moved at a uniform rate and the fibre images were detected on photomultiplier tubes and converted into electrical pulses the duration of which was proportional to the area of the fibres, and hence to the transverse dimension. Since the fibre alignment was not always complete, the image of the fibre was split by use of a pair of prisms. The two images were scanned by separate photomultiplier tubes, which were parallel to and in line with the length of the fibres. This arrangement caused a fibre that was parallel to the slit to cause an electrical pulse from both photomultiplier tubes at the same time. However a fibre image that was not parallel to the slit caused an electrical output from one photomultiplier, which was out of phase with the pulse from the second tube. By this means the instrument discriminated against fibres that were not correctly aligned.

Hourihan, Terrill, Neil and Mackey (1970) reported on the application of this instrument to the measurement of the diameter of wool tops.

The instrument was calibrated with IWS tops. Its performance against the projection microscope method for a series of validation tops is illustrated in the following table.

The instrument had a clear diameter dependent bias, being coarser for the fine wools and becoming progressively finer as the diameter increased. The standard deviation was considerably larger. The authors recognised this but suggested the instrument was at least satisfactory for ranking results, and provided a much more rapid measurement than the projection microscope method, and thereby providing a useful tool for quantitative geneticists for ranking animals.

Thorn Bendix in co-operation with the Textile Department at the University of Leeds developed an instrument called the Fibre Diameter Analyser in 1969 (Anon. 1969). Details on this device are scanty but the principle was relatively simple.

Figure 2: A Schematic of the Fibre Diameter Analyser developed by the University of Leeds in 1969.

A sample batch of 300 fibres was mounted on the perimeter of a circular sample holder, parallel to the axis. The holder was rotated at 10 revolutions per minute so that each fibre in turn passed between stabilised light source and a photo-detector. The electrical pulses obtained varied in magnitude depending upon the diameter of the fibre. It was claimed that the instrument had a range of 10 to 70 microns, and a precision better than 0.5 microns. This instrument has since vanished into obscurity.

Leaping Forward with Laser Optics (1970-1983)

non-uniform along the filaments. Lynch and Thomas concluded that the optical diffraction profiles of single fibres offered the possibility of measuring continuously the variation in diameter and cross section along the length of the fibres at a length-interval resolution of at least 0.5 mm and possibly as great as a few diameters. They foreshadowed that although the diffraction profiles of natural fibres could not be used in conjunction with some of the simpler scattering equations to give absolute values of diameter and cross-section, an empirical calibration of the scattering angle of the first diffraction minimum against diameter, as measured by some other system, could be made with some precision.

Lynch and Michie (1976) described the design principles, construction and operation of an instrument designed for the rapid automatic measurement of fibre fineness distribution, and of course mean diameter. The instrument was based on the electro-optical measurement of the amount of light scattered from a directed beam generated by a laser by fibre snippets. The fibre snippets were transported through the beam dispersed in a moving liquid.

Lynch and Michie’s paper is a very carefully constructed description of the instrument and identifies the critical features in the design. These included:

• The geometry of the beam of light, including its shape and its area;
• The presentation of the fibres to the beam;
• The characteristics of the liquid used to transport the fibres;
• The orientation of the fibres in the beam;
• The discrimination of fibres that are suitable to measure from those that are unsuitable;
• The discrimination of non-fibrous material and fibre fragments from actual fibres;
• The discrimination of signals produced from a single fibre from those produced by multiple fibres;
• The stability of the electro-optics;
• The precise control of the temperature of the liquid transporting the fibres;
• The selection of the transporting liquid; and
• The desirability to calibrate the instrument to be in as close as possible agreement with the Projection Microscope.

Schematic of the method of fibre snippet presentation for measurement. The snippets are transported in a liquid through a glass conduit of square section through which the light beam passes.

The instrument utilised the unique correlation between the amount of light scattered from a directed light beam by a fibre and the fineness of the fibre. By using a very low angle of detection, the instrument avoided any problems arising from the fact that wool fibre is non-absorbing of light and is irregular in geometrical and material properties. By using a beam of light of circular symmetry and causing the fibre to intersect the beam at right angles to the direction of the beam, a unique situation occurred in transit when the maximum amount of light was scattered by the beam. This occurred when the axis of the fibre and the axis of the beam intersected, and because the beam was circular, it was independent of the orientation of the fibre in the plane of intersection. For a parallel beam of uniform irradiance the amount of light scattered is clearly proportional to the area of the projected shadow or image of the fibre. If this area is taken at a unique point of passage when the fibre is intersecting the beam’s axis then the projection area or area of light extinction is maximum and defined by the equation: 

There is an almost linear relationship between A and d up to a value of d =D/6. By taking account of this and also allowing for curvature due to crimp the instrument was designed with a beam of 200 micrometres diameter. This was generated by directing the laser beam through a 200 micrometre pinhole and setting a circular 2 mm aperture in the far field of the 200 micrometre pinhole to accept only the central lobe of the resultant diffraction pattern. This produced a slowly diverging diffraction limited beam, of circular cross-section. The irradiance of the beam decreased radially. To compensate for this, the plane of intersection for the scattering of the beam by the fibres was chosen to be located at a point in the far field of the pinhole where the diameter was approximately 300 micrometres.

The fibres were presented to the beam dispersed as a liquid slurry, which was channelled through a squared-sectioned, fused glass conduit which confined the slurry to a laminar region 2 mm deep intersecting the beam.

With this arrangement not all fibres fully intersected the beam. To discriminate against such events, the instrument used a split circular photo detector. If the signals from the two semi-circular detectors did not match then the event was not recorded as a valid fibre. This method of detection also enabled the rejection of signals produced by contaminating particles. The possibility of multiple fibres traversing the detector simultaneously was minimised by ensuring that the slurry concentration was low and a maximum count rate was not exceeded. In the event that two fibres crossed the detector simultaneously, the electronics was set to detect the multiple peaks above a base threshold that such an event produced, and then to reject the event.

The stability of the laser power was critical to the instrument. This was achieved by a feedback loop whereby the actual measurement beam was sampled by a reference photodiode and the detected intensity used to regulate the laser discharge current. This regulation maintained long term stability and a suitable baseline reference.

Precise temperature control was required to minimise fluctuations of the optical properties of the transporting fluid and the air through which the laser beam travelled. Temperature fluctuations within discrete domains of either fluid can cause excessive noise in the instrument. Laser beams are particularly sensitive to this effect. The precise control of temperature required the instrument to be located in a temperaturecontrolled cabinet, and also required the transporting liquid to be recirculated. Thus a filter system was incorporated to remove the fibre snippets after they passed through the measurement cell.

The liquid chosen to transport the snippets was required to have a number of properties. It needed to:

• be capable of dispersing the snippets;
• have as large as possible refractive index difference from that of wool;
• have a low coefficient of refractive index change with temperature;
• be transparent to the light beam;
• not cause undesirable changes in the physical properties of the wool such as unpredictable swelling of the fibres;
• be inactive chemically with wool;
• have low viscosity at room temperature; and
• be non-toxic.

Isopropanol was chosen because of all the available liquids it best fitted these criteria.

Lynch and Michie reported a preliminary evaluation of the instrument. They used 1.5 mm snippets because this length was long enough to ensure a good probability of the fibres fully intersecting the beam and not too long to cause entanglements and resultant blockages in the circulating system and conduits. They observed that preparation systems that caused felting of the fibre should be avoided. Felting caused measurement error due to the significant number of fibre snippets that remained in contact during measurement.

These authors finally concluded that the instrument was potentially effective for the measurement of fibre fineness. That is:

• it was sufficiently reproducible for repeated measurement on single samples;
• it was sufficiently reproducible for a series of subsamples drawn from a well blended wool; and
• correspondence with fibre fineness measurements using Projection Microscope appeared to be statistically significant.

Lynch and Michie lamented the dearth of suitably calibrated reference wools, pointing out that these were critical for the optimum calibration of the instrument, and they foreshadowed the preparation of such reference tops by the CSIRO.

The instrument described by Lynch and Michie and physically constructed by CSIRO is now known as the Fibre Fineness Distribution Analyser (FFDA) or alternatively as the Fibre Distribution Analyser (FDA).

Irvine and Lunney (1979) described a procedure for calibrating this instrument in such a way as to bring its readings into conformity with the Projection Microscope method. The justification Irvine and Lunney presented for a physical calibration against wool fibres of known distribution was based on two factors:

• The physical design of the instrument was such that it was impracticable to develop a simple calibration function that onverted signals generated from fibre events directly into diameter. Firstly, the laser beam varied radially in its intensity. Secondly, the way the light was scattered by wool depends in a complex way on the bulk geometrical and material properties of the fibre. Thirdly, the size of the 200 micrometre pinhole was difficult to standardise so it was probable that each instrument would have a slightly different calibration.
  
• The aim of the calibration was to cause measurements by the instrument to conform as closely as possible to the Projection Microscope, which was then and still is the accepted international standard for establishing the distribution of fineness in wool.

Lunney and Irvine also described a number of factors affecting measurements on wool top produced by this instrument. These factors influenced the choice of conditions under which calibration and measurement was carried out, and included:

• Method of cutting fibre snippets;
• Proportion of snippets actually counted;
• Physical slice length compared with actual snippet length;
• Water content of the isopropanol;
• Effect of snippet length on the measured diameter;
• The actual snippet length to be used for calibration and measurement;
• The rate at which snippets were counted; and
• The number of coincidence counts.

These authors recommended calibrating the instrument with 2% water content in the isopropanol. They also recommended preparing snippets for calibrating the instrument in the same manner as snippets would be prepared from unknown samples prior to measurement. They concluded that both calibration and measurement required some control of the count rate, and suggested a maximum raw count rate of 20 counts per second. Snippets less than 0.4 mm were found to give very fine results. For longer snippets (up to 3 mm), a 21 micron top was not affected by snippet length, while the mean of a 32 micron top reduced by about 0.8 micrometres for every added 1 mm in length.

In a subsequent report, Lunney and Irvine (1982) revised their earlier recommendation for water content in the isopropanol, increasing the recommended level to 8.5%. This followed practical experience that the moisture content gradually increased with time and the 2% limit required too frequent adjustment. They also adopted a higher count rate (50 counts per second) than previously recommended.

Dr Leo Lynch, one of the developers of the FDA (FFDA) is now retired. He is pictured here, fifth from the left, during a visit to AWTA Ltd's Research and Development Division in Sydney on 11th Janurary 2001, in the workshop where the modern Sirolan Laserscan is assembled.

¹   This is but one principle that can be applied, and is described here because it is the principle behind modern photometric instruments.  However earlier instruments utilised the birefringent properties of wool and polarised light to create an image that could cause a response in a photo-dector.

²   Division of Textile Physics, CSIRO, Ryde, NSW, Australia (the division has now been relocated to Belmont, Victoria)