Each type of bakery product has a typical form and texture, which is recognised by the consumer and a narrow range of textures, which define its nature. Any significant deviation from the ideal texture characteristics of a product type will often be considered as a loss of quality. Therefore, manufacturers aim to manufacture products whose texture remains within the specific limits. In order to achieve this aim they must be able to assess their products at the factory and during storage trials to assure a consistent level of quality for the consumer. Quality assessments can be carried out by trained taste panels, who would examine suitable numbers of product samples after every production run. However, such procedures are labour intensive and costly even with a limited number of sensory attributes.

Therefore, it has been the practice in the baking industry since the 1930s to attempt to develop objective instrumental methods for assessing product texture. This review covers some of the methods, which have been found to be most useful for our research and contract studies at CCFRA.

There is another aspect of texture measurement, which is equally important and well understood in the industry. Bakery products must not only be attractive to the consumer in their eating quality, but they must have sufficient strength and resilience in their structures to survive the handling, packaging and transport processes. One of the standard tests on pies and cakes is the road trial in which the products are subjected to a long transport journey in a truck to detect any fragility in their structure.

Therefore, objective testing methods have a second role in which their measurements can be used to assess the robust nature of the product.

In this review I shall divide the methods that have been used for the assessment of bakery products into two groups. The first group will cover the hard brittle foods such as biscuits and pastries and the second group will look at the soft moist foods such as breads and cakes.

Hard brittle foods: biscuits, pastries and snacks
It is difficult to assess the texture of hard brittle foods with instrumental methods because of their variable shape and texture, and their fragility. The earliest approaches involved either compression to crush the biscuits and snacks, or three point bending and snapping tests. These methods gave useful information to compare with the sensory score for hardness at the first bite into the product. However, the methods were not very reproducible for batches from different bakes. It was often difficult to maintain a constant area of contact between the probes and the products and the fracture patterns in the products tended to be irregular.

A method for measuring the hardness of semi-sweet biscuits was developed which monitored the torque input to a circular saw as it cut a path through a stack of biscuits. This method gave a good correlation with sensory scores for hardness and was used in the biscuit industry.

The most successful methods yet developed for the assessment of brittle foods are based on penetrometry techniques. In the 1980s CCFRA developed penetrometry methods for the assessment of pastries and extruded snack foods with an automated testing machine. In both cases a small rod < 10mm diameter was driven into the samples and a force-distance graph recorded on paper charts. This trace was interpreted to give information on the hardness of first bite and the size of layers and cells within the product structure, through which the probe passed.

The technique of penetrometry was developed further for biscuits1 and extruded foods2 with computer aided compressimeters. For the assessment of biscuits conical probes with half angles of 10 to 15O (Figure1)

Figure 1
Penetrometer with small angle cone for assessing biscuit texture

were driven into the products to record a force distance graph. Moreover, the graph was recorded digitally at 200-400 points/s so that individual peaks on the force distance graph (Figure 2)

Figure 2
Force-distance graph for a biscuit on cone penetrometer

could be measured. After full analysis they could be described in terms of the breaking forces required to remove small structures in the biscuit. Analysis of the graphs was simplified by using suitable computer programmes to provide measures of hardness, fractures per mm and a distribution graph for fractures of particular size range in terms of breaking force.

Comparisons with scores from a trained sensory panel for short dough biscuits at CCFRA showed that the cone penetrometry could be used to evaluate the first bite attributes, such as hardness, crunchiness and crumbliness. This confirmed the earlier work1 and set the method up in a form, which could be easily used by laboratory staff for all types of hard sweet and short dough biscuits.

The penetrometry method for extruded products developed at CCFRA was also converted to a digital method (Figure 3)

Figure 3
Force-distance graph for penetrometry of a snack product with a 4mm rod

and a computer programme has been devised to analyse the fractures shown on the graph. The recent studies2 have confirmed that the analysis of force-distance penetration graphs provides information on the eating quality of the extruded products. At CCFRA we have begun to apply this technique to snack biscuits and cream crackers, whose cell wall structures are more similar to extrudates than hard sweet or short dough biscuits.

The use of penetrometry with pastry has been carried out at CCFRA with both short crust sweet pastry for fruit pies and boiled water pastry for meat pies. A small rod of <10mm diameter is used and the force distance curve is recorded digitally so that it can be analysed automatically. The typical graph show a simpler pattern than extrudates or biscuits with a simple peak force for the outer layer of the pastry followed by smoother transitions through the softer inner layer. In both cases there is a good correlation between sensory assessment of pastry hardness and the objective measurement.

Soft Moist Foods: breads and cakes
The use of objective instrumental methods for the assessment of bakery products began with bread and many hundreds of papers have been published of the subject. In the 1960s CCFRA developed the "Cone Indenter"3 to measure crumb softness and this instrument was shown to give a very good correlation with the taste panels assessment of both bread and cake staleness. This machine was superseded by the arrival of automated testing machines equipped to measure forces in either compression or tension. The reproducibility of the Cone indenter was slightly superior but the new machines with their availability, ease of calibration and direct evaluation of the elastic modulus of bread led to their widespread usage. Compressimetry methods for bread and cake were adapted further at CCFRA in the 1980s to incorporate density corrections and to apply a two-stage compression technique (Figure 4). This was adapted from the Texture Profile Analysis method of Bourne5 in which a food is compressed to give a 25% strain and then released from its load, before immediately carrying out a second compression. For bread the TPA method gives a value for firmness and the recovery or resilience of the crumb after the 1st compression (Figure 4).

Figure 4
Texture profile analyse graph for two successive compressions of cake crumb

The TPA method has been used widely and was adopted as a standard by the AACC without the correction for density. CCFRA has continued to improve the method by introducing a computer analysis to extract the critical factors and record them automatically together with statistical information. In recent studies on the use of maltogenic amylases to prolong the shelf-life of bread, there have been good correlations between the TPA method and taste panel attributes for the 1st bite of hardness by taste and touch (Figure 4).

Another very useful method for examining bread was developed in 1970 by Bishop and Wren. This was designed to reproduce the squeezing action of a consumer to test the firmness of a sliced loaf. Compressimetry with a short metal bar was carried out on the side of the loaf to record a force measurement. The details of the test were adjusted until they gave a good correlation in triangular tests with the sensory results for normal laboratory staff.

Summary and view on the future
The current methods used in the industry are very useful in helping to maintain quality standards. However, they are limited in their ability to assess bakery products compared with human sensory facilities. At the moment there is a reasonable coverage of methods, which are correlated with 1st bite attributes, but no satisfactory method to cover the second stage of eating and sensory appreciation of a product, the mastication of products, where there are variable factors related to the addition of saliva.

Studies with mechanical testing machines which inject a solution of enzymes have been attempted but have met with limited success. Some equipment has been developed to record the electrical impulses from face muscles and from sensors located on a tasterÕs teeth but these methods are complex and difficult to calibrate. Their best use is to provide information to help in the design of new instrumental methods. Currently CCFRA is carrying out research studies on the best of the existing methods available, to refine them and develop national standards.

In the second area of interest concerning the ability of bakery products to withstand the rigors of handling and transport there are a number of methods available for mechanical testing under different conditions. These include impact testing and low deformation repetitive strain instruments. The latter type of equipment include parallel plate and cone and plate rheometers and other special devices equipped to apply sinusoidal stresses.

1. Goullieux, A., Allaf, K., and Bouvier, J-M. (1995) Sciences des aliments, 15, 3-18
2. Van Hecke, E., Allaf, K. and Bouvier, J.M. (1998) J. Texture Studies, 29, 617-632.
3. Axford, D.W.E., Colwell, K.H, Cornford, S. J and., Elton G.A.H (1968), J. Sci Food Agriculture, 19, 95-101.
4. Bourne, M.C. (1978) Food Technology, 72, July, 62-66.
5. Bishop, E. E. and Wren J.J. (1971) Compressimetry of bread, J. Food Technol, 6, 409