Lipolysis in milk and dairy products: a research journey

In this paper, I describe my journey through a field of research in which I have been involved for some years - lipolysis in milk and dairy products. While I call it my journey, I have had many fellow travellers who have helped me along the way. These have been my research colleagues and collaborators, and, since I joined the University of Queensland, my students. The research has covered a variety of aspects but I have chosen to describe only a selection of these.

Factors affecting cane yield and commercial cane sugar in the Tully district

The Tully Sugar Mill has collected information about sugarcane supplied for crushing from every block in the mill district from 1970 to 1999. Data from 1988 to 1999 were analysed to understand the extent of the variation in cane yield per hectare and commercial cane sugar in the Tully mill area. The key factors influencing the variation in cane yield and commercial cane sugar in this commercial environment were identified and the variance components computed using a restricted maximum likelihood methodology. Cane yield was predominantly influenced by the year in which it was harvested, the month when the crop was ratooned (month of harvest in the previous year) and the farm of origin. These variables were relatively more important than variety, age of crop or crop class (plant crop, first ratoon through to fourth or older ratoons) and fallowing practice (fallow or ploughout-replant). The month-of-ratooning effect was relatively stable from year-to-year. Commercial cane sugar was influenced by the year of harvest, the month of harvest and their interaction, in that the influence of the month of harvest varied from year to year. Variety and farm differences were also significant but accounted for a much lower portion of the variation in commercial cane sugar. An empirical model was constructed from the key factors that influenced commercial cane sugar and cane yield to quantify their combined influence on sugar yield (t/ha). This may be used to assist mill personnel to predict their activities more accurately, for example to calculate the impact of a late finish to the current harvest season on the following year's crop.

Application of the Williams-Landel-Ferry model to the viscosity-temperature relationship of Australian honeys

The rheological behaviour of nine unprocessed Australian honeys was investigated for the applicability of the Williams-Landel-Ferry (WLF) model. The viscosity of the honeys was obtained over a range of shear rates (0.01-40 s(-1)) from 2degrees to 40 degreesC, and all the honeys exhibited Newtonian behaviour with viscosity reducing as the temperature was increased. The honeys with high moisture were of lower viscosity, The glass transition temperatures of the honeys, as measured with a differential scanning calorimeter (DSC), ranged from -40degrees to -46 degreesC, and four models (WLF. Arrhenius, Vogel-Tammann-Fulcher (VTF), and power-law) were investigated to describe the temperature dependence of the viscosity. The WLF was the most suitable and the correlation coefficient averaged 0.999 +/- 0.0013 as against 0.996 +/- 0.0042 for the Arrhenius model while the mean relative deviation modulus was 0-12% for the WLF model and 10-40% for the Arrhenius one. With the universal values for the WLF constants, the temperature dependence of the viscosity was badly predicted. From non-linear regression analysis, the constants of the WLF models for the honeys were obtained (C-1 = 13.7-21.1: C-2 = 55.9-118.7) and are different from the universal values. These WLF constants will be valuable for adequate modeling of the rheology of the honeys, and they can be used to assess the temperature sensitivity of the honeys. (C) 2002 Elsevier Science Ltd. All rights reserved.

Discrete element modelling of the influence of lifters on power draw of tumbling mills

Crushing and grinding are the most energy intensive part of the mineral recovery process. A major part of rock size reduction occurs in tumbling mills. Empirical models for the power draw of tumbling mills do not consider the effect of lifters. Discrete element modelling was used to investigate the effect of lifter condition on the power draw of tumbling mill. Results obtained with PFC3D code show that lifter condition will have a significant influence on the power draw and on the mode of energy consumption in the mill. Relatively high lifters will consume less power than low lifters, under otherwise identical conditions. The fraction of the power that will be consumed as friction will increase as the height of the lifters decreases. This will result in less power being used for high intensity comminution caused by the impacts. The fraction of the power that will be used to overcome frictional resistance is determined by the material's coefficient of friction. Based on the modelled results, it appears that the effective coefficient of friction for in situ mill is close to 0.1. (C) 2003 Elsevier Science Ltd. All rights reserved.

Applying discrete element modelling to vertical and horizontal shaft impact crushers

The PFC3D (particle flow code) that models the movement and interaction of particles by the DEM techniques was employed to simulate the particle movement and to calculate the velocity and energy distribution of collision in two types of impact crusher: the Canica vertical shaft crusher and the BJD horizontal shaft swing hammer mill. The distribution of collision energies was then converted into a product size distribution for a particular ore type using JKMRC impact breakage test data. Experimental data of the Canica VSI crusher treating quarry and the BJD hammer mill treating coal were used to verify the DEM simulation results. Upon the DEM procedures being validated, a detailed simulation study was conducted to investigate the effects of the machine design and operational conditions on velocity and energy distributions of collision inside the milling chamber and on the particle breakage behaviour. (C) 2003 Elsevier Ltd. All rights reserved.

The JK three-product cyclone - performance and potential applications

In spite of their wide application in comminution circuits, hydrocyclones have at least one significant disadvantage in that their operation inherently tends to return the fine denser liberated minerals to the grinding mill. This results in unnecessary overgrinding which adds to the milling cost and can adversely affect the efficiency of downstream processes. In an attempt to solve this problem, a three-product cyclone has been developed at the Julius Kruttschnitt Mineral Research Centre (JKMRC) to generate a second overflow in which the fine dense liberated minerals can be selectively concentrated for further treatment. In this paper, the design and operation of the three-product cyclone are described. The influence of the length of the second vortex finder on the performance of a 150-mm unit treating a mixture of magnetite and silica is investigated. Conventional cyclone tests were also conducted under similar conditions. Using the operational performance data of the three-product and conventional cyclones, it is shown that by optimising the length of the second vortex finder, the amount of fine dense mineral particles that reports to the three-product cyclone underflow can be reduced. In addition, the three-product cyclone can be used to generate middlings stream that may be more suitable for flash flotation than the conventional cyclone underflow, or alternatively, could be classified with a microscreen to separate the valuables from the gangue. At the same time, a fines stream having similar properties to those of the conventional overflow can be obtained. Hence, if the middlings stream was used as feed for flash flotation or microscreening, the fines stream could be used in lieu of the conventional overflow without compromising the feed requirements for the conventional flotation circuit. Some of the other potential applications of the new cyclone are described. (C) 2003 Elsevier Science B.V. All rights reserved.

Slurry flow in mills: grate-pulp lifter discharge systems (Part 2)

Pulp lifters, also known, as pan lifters are an integral part of the majority of autogenous (AG), semi-autogenous (SAG) and grate discharge ball mills. The performance of the pulp lifters in conjunction with grate design determines the ultimate flow capacity of these mills. Although the function of the pulp lifters is simply to transport the slurry passed through the discharge grate into the discharge trunnion, their performance depends on their design as well as that of the grate and operating conditions such as mill speed and charge level. However, little or no work has been reported on the performance of grate-pulp lifter assemblies and in particular the influence of pulp lifter design on slurry transport. Ideally, the discharge rate through a grate-pulp lifter assembly should be equal to the discharge rate through at a given mill hold-up. However, the results obtained have shown that conventional pulp lifter designs cause considerable restrictions to flow resulting in reduced flow capacity. In this second of a two-part series of papers the performance of conventional pulp lifters (radial and spiral designs) is described and is based on extensive test work carried out in a I m diameter pilot SAG mill. (C) 2003 Elsevier Science Ltd. All rights reserved.

Slurry flow in mills: grate-only discharge mechanism (Part-1)

Discharge grates play an important role in determining the performance of autogenous, semi-autogenous and grate discharge ball mills. The flow capacity (grinding capacity) of these mills is strongly influenced by the discharge grate design-open area and position of apertures, as well as the performance of the pulp lifters. As mill sizes have progressively increased and closed-circuiting has become more popular the importance of grate and pulp lifter design has grown. Unfortunately very few studies have concentrated on this aspect of mill performance. To remedy this a series of laboratory and pilot-scale tests were undertaken to study both the performance of grates on their own and in conjunction with pulp lifters. In this first paper of a two-part series the results from the grate-only experiments are presented and discussed, whilst the performance of the grate-pulp-lifter system is covered in the second paper. The results from the grate-only experiments have shown that the build-up of slurry (hold-up) inside the mill starts from the shoulder of the charge, while the toe position of the slurry progressively moves towards the toe of the charge with increasing flowrate. Besides grate design (open area and position of apertures), charge volume and mill speed were also found to have a strong influence on mill hold-up and interact with grate design variables. (C) 2003 Elsevier Science Ltd. All rights reserved.

Discrete element modelling of power draw of tumbling mills

The power required to operate large mills is typically 5-10 MW. Hence, optimisation of power consumption will have a significant impact on overall economic performance and environmental impact. Power draw modelling results using the discrete element code PFC3D have been compared with results derived from the widely used empirical Model of Morrell. This is achieved by calculating the power draw for a range of operating conditions for constant mill size and fill factor using two modelling approaches. fThe discrete element modelling results show that, apart from density, selection of the appropriate material damping ratio is critical for the accuracy of modelling of the mill power draw. The relative insensitivity of the power draw to the material stiffness allows selection of moderate stiffness values, which result in acceptable computation time. The results obtained confirm that modelling of the power draw for a vertical slice of the mill, of thickness 20% of the mill length, is a reliable substitute for modelling the full mill. The power draw predictions from PFC3D show good agreement with those obtained using the empirical model. Due to its inherent flexibility, power draw modelling using PFC3D appears to be a viable and attractive alternative to empirical models where necessary code and computer power are available.

Asymmetric blasting: a rock mass dependent blast design method

Blasting has been the most frequently used method for rock breakage since black powder was first used to fragment rocks, more than two hundred years ago. This paper is an attempt to reassess standard design techniques used in blasting by providing an alternative approach to blast design. The new approach has been termed asymmetric blasting. Based on providing real time rock recognition through the capacity of measurement while drilling (MWD) techniques, asymmetric blasting is an approach to deal with rock properties as they occur in nature, i.e., randomly and asymmetrically spatially distributed. It is well accepted that performance of basic mining operations, such as excavation and crushing rely on a broken rock mass which has been pre conditioned by the blast. By pre-conditioned we mean well fragmented, sufficiently loose and with adequate muckpile profile. These muckpile characteristics affect loading and hauling [1]. The influence of blasting does not end there. Under the Mine to Mill paradigm, blasting has a significant leverage on downstream operations such as crushing and milling. There is a body of evidence that blasting affects mineral liberation [2]. Thus, the importance of blasting has increased from simply fragmenting and loosing the rock mass, to a broader role that encompasses many aspects of mining, which affects the cost of the end product. A new approach is proposed in this paper which facilitates this trend 'to treat non-homogeneous media (rock mass) in a non-homogeneous manner (an asymmetrical pattern) in order to achieve an optimal result (in terms of muckpile size distribution).' It is postulated there are no logical reasons (besides the current lack of means to infer rock mass properties in the blind zones of the bench and onsite precedents) for drilling a regular blast pattern over a rock mass that is inherently heterogeneous. Real and theoretical examples of such a method are presented.

An energy-based model for swing hammer mills

An energy-based swing hammer mill model has been developed for coke oven feed preparation. it comprises a mechanistic power model to determine the dynamic internal recirculation and a perfect mixing mill model with a dual-classification function to mimic the operations of crusher and screen. The model parameters were calibrated using a pilot-scale swing hammer mill at various operating conditions. The effects of the underscreen configurations and the feed sizes on hammer mill operations were demonstrated through the fitted model parameters. Relationships between the model parameters and the machine configurations were established. The model was validated using the independent experimental data of single lithotype coal tests with the same BJD pilot-scale hammer mill and full operation audit data of an industrial hammer mill. The outcome of the energy-based swing hammer mill model is the capability to simulate the impact of changing blends of coal or mill configurations and operating conditions on product size distribution. Alternatively, the model can be used to select the machine settings required to achieve a desired product. (C) 2003 Elsevier Science B.V. All rights reserved.

Critical comparison of Julius Kruttschnitt Mineral Research Centre (JKMRC) blast fragmentation models

Blast fragmentation can have a significant impact on the profitability of a mine. An optimum run of mine (ROM) size distribution is required to maximise the performance of downstream processes. If this fragmentation size distribution can be modelled and controlled, the operation will have made a significant advancement towards improving its performance. Blast fragmentation modelling is an important step in Mine to Mill™ optimisation. It allows the estimation of blast fragmentation distributions for a number of different rock mass, blast geometry, and explosive parameters. These distributions can then be modelled in downstream mining and milling processes to determine the optimum blast design. When a blast hole is detonated rock breakage occurs in two different stress regions - compressive and tensile. In the-first region, compressive stress waves form a 'crushed zone' directly adjacent to the blast hole. The second region, termed the 'cracked zone', occurs outside the crush one. The widely used Kuz-Ram model does not recognise these two blast regions. In the Kuz-Ram model the mean fragment size from the blast is approximated and is then used to estimate the remaining size distribution. Experience has shown that this model predicts the coarse end reasonably accurately, but it can significantly underestimate the amount of fines generated. As part of the Australian Mineral Industries Research Association (AMIRA) P483A Mine to Mill™ project, the Two-Component Model (TCM) and Crush Zone Model (CZM), developed by the Julius Kruttschnitt Mineral Research Centre (JKMRC), were compared and evaluated to measured ROM fragmentation distributions. An important criteria for this comparison was the variation of model results from measured ROM in the-fine to intermediate section (1-100 mm) of the fragmentation curve. This region of the distribution is important for Mine to Mill™ optimisation. The comparison of modelled and Split ROM fragmentation distributions has been conducted in harder ores (UCS greater than 80 MPa). Further work involves modelling softer ores. The comparisons will be continued with future site surveys to increase confidence in the comparison of the CZM and TCM to Split results. Stochastic fragmentation modelling will then be conducted to take into account variation of input parameters. A window of possible fragmentation distributions can be compared to those obtained by Split . Following this work, an improved fragmentation model will be developed in response to these findings.

Amylases in enzyme detergents and their effects on a simulated long-life dairy dessert

Enzyme detergents used in the food industry contain proteinase as the major enzyme but amylase may be present, either by design or inadvertently. Three commercial enzyme detergents and 3 enzyme preparations used in detergents were assayed for alpha-amylase activity by the Ceralpha method using the Megazyme kits. The amylase activities of the detergents varied from 3.2x 10(-6) to 32x 10(-6) mumoles ml(-1) h(-1) while the enzyme preparations had much higher activities ranging from 0.05 to 8.06 mumoles ml(-1) h(-1). When added aseptically to a simulated dairy dessert (2% starch solution) and stored for 42 days, the enzyme detergents caused an increase in viscosity; enzyme preparations at low concentrations caused an initial increase in viscosity followed by a decrease; and enzyme preparations at high concentrations caused an immediate decrease in viscosity. The increase in viscosity corresponded to formation of a distinct network of starch granules while the decrease in viscosity was characterised by a marked decrease in size of the granules and little or no network of granules. Decreases in viscosity corresponded to increases in reducing sugars but samples which increased in viscosity showed no measurable reducing sugars. The amylase activity in all sources was destroyed by heating at 75degreesC for 15 min at pH 1.8.

Heat resistance of Bacillus spores when adhered to stainless steel and its relationship to spore hydrophobicity

Twenty-one strains of Bacillus (10 B. stearothermophilus, 3 B. cereus, and 8 B. licheniformis strains) were assayed for spore surface hydrophobicity on the basis of three measures: contact angle measurement (CAM), microbial adhesion to hydrocarbons (MATH), and hydrophobic interaction chromatography (HIC). On the basis of the spore surface characteristics obtained from these assays, along with data on the heat resistance of these spores in water, eight strains of Bacillus (three B. stearothermophilus, three B. cereus, and two B. licheniformis strains) either suspended in water or adhering to stainless steel were exposed to sublethal heat treatments at 90 to 110degreesC to determine heat resistance (D-value). Significant increases in heat resistance (ranging from 3 to 400%) were observed for the eight strains adhering to stainless steel. No significant correlation was found between these heat resistance increases and spore surface characteristics as determined by the three hydrophobicity assays. There was a significant positive correlation between the hydrophobicity data obtained by the MATH assay and those obtained by the HIC assay, but these data did not correlate with those obtained by the CAM assay.