WWW.WAS.ORG • WORLD AQUACULTURE • MARCH 2025 27 Trial Design The study was designed to compare the three extrusion systems: • ‘AFX’ (Advanced Twin-Screw Technology): Features innovative screw elements designed for enhanced wear resistance and flexibility. • ‘TX’ (Traditional Twin-Screw Technology): Represents the standard twin-screw design used in many current, and often more demanding, applications. • SX (Single-Screw Technology): A simpler, robust extruder with a single screw. This represents what is often and historically used in many feed extrusion applications. Comparative production runs were conducted using a standardized salmon feed formulation (Table 1) over five different sets of processing conditions on three extrusion platforms (AFX, TX, SX). All runs utilized similar setup for feeding and preconditioning equipment. The utilization of five different sets of processing conditions was to mimic resilience towards operator choice as well as to test for raw material flexibility (e.g. different ‘intrinsic levels’ of moisture and fat). The different run conditions tested on each of the three extrusion platforms were: 1. High SME/STE* ratio. This run condition will utilize mechanical energy more than thermal energy to cook, bind and shape the feed pellets. 2. Low SME/STE. Higher levels of steam to the conditioner and to the extrusion barrel than in ‘1’. 3. Low water addition [water inclusion dropped by ~4% (of dry feed rate)] 4. Low oil addition (0%) 5. High oil addition (4%) *SME=Specific Mechanical Energy, STE=Specific Thermal Energy. Whilst all runs fall into the category of stress testing the ‘process window,’ e.g. as different extrusion operator experience and approaches, particularly runs 4 and 5 are examples of testing raw material flexibility, mimicking different intrinsic lipid levels of fat in the recipe. Common to all runs was that a product bulk density off the extruder of 400 g/L was targeted. To achieve as good-of quality products as possible and density on target, any processing parameters could be altered, aside from the parameters that specify the actual characteristics of the particular run condition, for example, in run ‘3,’ water inclusion was preset. Production rate was approximately 700 kg/h across all runs. Identical upstream conditions were maintained and all runs utilized the same die with 6.5 mm holes, targeting a ~9 mm grower feed. Temperatures in the dryer following the extrusion step were maintained constant at ~112 °C. Retention time in the dryer varied from 27-32 minutes to obtain a final moisture target of ~9%, accounting for different process water and steam settings in the extruder. The bulk density evaluation was done immediately as the product exited the extruder. Once the quality and density of each run was optimized, the run was logged and product samples were taken from the dryer outlet, corresponding to the retention time in the dryer, when dryer moisture was also on target. The evaluation of pellet physical quality attributes for each run was carried out on the dried pellets: • Pellet diameter 1 + 2: 30 pellets from each run were measured using a digital caliper on the shorter (1) as well as on the longer side (2). • Circularity: Diameter 1 / Diameter 2. • Surface Area/Volume: 2*(D1+D2+4L)/(L*(D1+D2)) • Pellet length • Durability: Assessed using tumbler box method to gauge resistance to fragmentation and abrasion (ASAE, 1991). • Bulk Density (ρ): The measured pellet density across different runs. • Oil uptake, vacuum (%): Measuring the amount of oil in pct that pellets were able to absorb and retain under vacuum @ 200 mbar for 2 minutes, with a 45 sec. release of vacuum. Results The first objective of the trial was to assess the ability to reach the bulk density target across each of the 5 different processing conditions. Figure 1 indicates that runs on SX were able to get to target density, except for the lower and higher oil addition runs ‘4’ and ‘5.’ Runs on the TX were able to get to target density except the higher oil inclusion. AFX was generally able to achieve target densities across the different run conditions. It is observed that the water levels used across the AFX runs varied significantly compared to those in the TX and SX run series, cf. Figure 2. This variation can be explained by AFX’s improved ability to respond to and utilize moisture as a process variable, ensuring density while maintaining a consistently high-quality product. As expected, water levels were higher when oil inclusion in the diet was low and vice versa. This adjustment compensates for changes in melt viscosity within the extrusion barrel, maintaining die pressures and physicochemical properties necessary for proper pellet expansion and achieving desired bulk density. Water inclusion on SX for the low oil inclusion run (‘4’) was also high, but here this compensation failed to TABLE 1. A standardized ‘salmon feed’ formulation used across all runs and extrusion platforms. Salmon feed recipe in benchmark study Meal (%) Finished product (%) (if coated) Whole wheat 22 14 Vital Wheat Gluten 8 5 Soy Protein Concentrate 47 30 Fishmeal 23 14 Fish oil (in coater- not applied here) 0 25 Rapeseed oil (in coater- not applied here 0 12 (CONTINUED ON PAGE 28)
RkJQdWJsaXNoZXIy MjExNDY=