NEW POLYMER CHEMISTRY FOR REFINERY LIQUOR DECOLOURISATION

Transcription

1 REFEREED PAPER NEW POLYMER CHEMISTRY FOR REFINERY LIQUOR DECOLOURISATION GODDARD S AND VAN ZYL M Buckman Africa, Buckman Boulevard Hammarsdale, KwaZulu-Natal South Africa P.O. Box 591, Hammarsdale, 3700 smgoddard@buckman.com mjvanzyl@buckman.com Abstract A new colour precipitant was investigated in the removal of visual colour in sugar refining processes. The results of laboratory evaluations and a short plant trial are presented. Laboratory investigations have indicated that a colour reduction in the region of 30 to 55% is achievable using the enhanced performance chemistry of the colour precipitant Bulab Bulab 5154 facilitated a higher colour removal than that achieved with polyamine chemistry colour precipitants conventionally used to assist in the refinery colour removal process. Bulab 5154 can be used as a process aid to consistently achieve final sugar colour targets, as well as providing additional decolourisation during periods of high colour loading. Laboratory investigations were conducted at two refineries using different refining processes to facilitate colour removal. Refinery A uses a carbonatation followed by sulphitation process and Refinery B uses a phosphatation followed by ion exchange process. Bulab 5154 facilitated excellent colour removal in liquor from both refining processes. Keywords: colour precipitant, sugar refinery, refinery colour removal, decolourisation Introduction With the increasing demand to achieve efficient and cost effective removal of colour in the production of high quality white sugar, a number of processes and chemicals have been investigated. Cationic flocculants such as polyamine chemistries are an example of decolourising agents used to precipitate colourants and other anionic impurities from sugar liquors (Bennett et al., 1971; Moodley, 1993). Cationic polymers can be used to bind and facilitate precipitation of colour bodies from refinery melt liquor by way of the negatively charged colour bodies being attracted to the positively charged polymer, thus forming an insoluble precipitate that can be easily removed from the refinery melt. In this work, polymers were evaluated under laboratory conditions, utilising refinery melt liquor that was subjected to various refining processes. These include (i) carbonatation where lime and carbon dioxide gas (recovered from boiler flue gas) form a calcium carbonate precipitate that absorbs colour particles, (ii) sulphitation, where sulphur dioxide gas is reacted with lime to produce a calcium sulphite precipitate which adsorbs to the colour bodies, and (iii) phosphatation whereby lime and phosphoric acid are reacted together to form a calcium phosphate precipitate, which absorbs and entraps the colour bodies. The colour bodies are 457

2 then removed either by filtration or by flotation in an air flotation unit using a high molecular weight anionic flocculant to enhance the separation of the solids from the melt liquor. The cationic polymeric decolourisation agent is used to enhance rather than replace the refining processes. Benefits include a greater performance efficiency as well as enhanced colour removal during periods of exceptionally high colour loading. Methodology Laboratory investigations The laboratory evaluation of the new generation colour precipitant was conducted on refinery melt from two sugar refineries which employ different refining processes, referred to as Refinery A which employs carbonatation followed by sulphitation, and Refinery B which uses phosphatation followed by ion exchange. The aim of the study was to compare Bulab 5154 to two polyamine chemicals and a dimethylamine polymer as a colour precipitant aid to enhance the removal of colour bodies from the refinery melt. A refinery melt sample of approximately 65 brix and at a temperature of 80 to 85 C was collected post-carbonatation but prior to the dosage point of the colour precipitant for Refinery A (Figure 1), whereas for Refinery B the sample was collected post-phosphatation but before addition of the anionic polyacrylamide and air flotation unit (Figure 2). Figure 1. Refinery A process flow diagram. 458

3 Figure 2. Refinery B process flow diagram. The liquor was divided into equal aliquots to which various products (Table 1) were dosed at 150 and 200 mg/l, which is within the normal dosage range used for colour precipitants in the refinery process. For Refinery B the laboratory investigation was conducted on different occasions where there was a large variation in the incoming raw melt colour. The colour precipitants were therefore evaluated at 150 mg/l in both high and low colour liquors. The colour precipitants were allowed a five minute contact time with the refinery melt to facilitate the interaction of the product with the colour bodies. In the case of the refinery melt from Refinery A (carbonatation/sulphitation) the ICUMSA 420 colour method (Anon, 2009) was used to determine the refinery melt colour directly after the five minute contact time with the colour precipitant. For Refinery B (phosphatation/ion exchange), after the five minute contact period with the colour precipitant, an anionic polyacrylamide polymer was dosed at a rate of 10 mg/l to the melt in order to facilitate agglomeration of the colour complexes. The clear melt was allowed to separate from the coagulated solids and was then collected for ICUMSA 420 colour determination. The ICUMSA 420 colour method entailed taking the 65 brix melt and diluting to 5 brix, filtering through a 0.45 µm membrane, followed by ph adjustment to ph 7 and determining the absorbance at 420 nm. A calculation then presented the colour reading. Table 1. Colour precipitants evaluated in the laboratory investigations. Colour precipitant Bulab 5031 Bulab 5552 Bulab 5154 Current refinery product Description Polyamine Polyamine Enhanced performance chemistry Dimethylamine polymer 459

4 Plant trial A short plant trial was run at Refinery A over a four day period. Bulab 5154 was dosed after the saturators but before the autofilter supply tank at between 120 and 150 mg/l during the course of the trial. ICUMSA 420 colour was measured on the raw melt and the liquor post carbonatation, colour precipitant addition and filtration as per the method used for the laboratory investigations. The percentage colour removal was then calculated and reported. The colours were monitored four hourly for one day prior to starting the Bulab 5154 trial, for two days during the course of the trial and for one day after Bulab 5154 addition was halted. The current refinery product was being applied during the one day pre- and one day post-trial at similar dosage rates. Results Refinery A (carbonatation/sulphitation process) laboratory evaluations Decolourisations of 44 and 56% were achieved with a dosage of 150 and 200 mg/l Bulab 5154, respectively. A significantly lower melt colour was achieved when a colour precipitant was used as compared to the control melt with no product dosage. Bulab 5154 outperformed the other colour precipitants evaluated (Table 2). Table 2. Decolourisation of refinery melt from Refinery A untreated melt liquor colour mg/l 200 mg/l Product ICUMSA 420 % ICUMSA 420 % colour Decolourisation colour Decolourisation Bulab % % Bulab % % Bulab % % Current refinery product % % Refinery B (phosphatation/ion exchange process) evaluations Evaluation of the colour precipitants in Refinery B melt indicated that at 150 mg/l colour precipitant good decolourisation was achieved, as seen in Figure 3. This is due to the mechanism of charge neutralisation whereby the cationic colour precipitant binds anionically charged colour bodies, thereby allowing their agglomeration and separation from the refinery liquor. If the product is dosed in excess, the solution will become cationically charged. This will result in charge repulsion and inhibit agglomeration and separation of colour bodies from the refinery liquor, and hence a reduction in colour removal. It is therefore important that the optimum dose is determined. 460

5 % Colour Reduction % Decolourisation Product Dosage (mg/l) Bulab 5031 Bulab 5552 Bulab 5154 Current Refinery product Figure 3. Decolourisation achieved with the various colour precipitants at dosages of 150 and 200 mg/l. The products were evaluated at 150 mg/l in both high and low colour refinery melt from Refinery B (Table 3). Bulab 5154 was found to produce excellent decolourisation in both the high and low colour refinery melts resulting in the region of 52 to 55% decolourisation, regardless of the initial melt colour. Table 3. Decolourisation of the refinery melt from Refinery B with 150 mg/l colour precipitant. High colour liquor Low colour liquor Product ICUMSA 420 colour % Decolourisation ICUMSA 420 colour % Decolourisation Bulab % % Bulab % % Bulab % % Current refinery product % % The tests were performed in triplicate, and the average results are recorded in Table

6 Table 4. Refinery A (carbonatation/sulphitation process) trial data. Day Time ICUMSA 420 colour % Colour Carbonatated Raw melt reduction liquor Comments 2:00 a.m :00 a.m :00 a.m Before Bulab 5154 Addition (average colour 2:00 p.m reduction of 34%) 6:00 p.m :00 p.m :00 a.m :00 a.m During the Addition of 10:00 a.m Bulab :00 a.m (average colour reduction 10:00 a.m of 43%) 11:30 a.m a.m After Bulab 5154 Addition p.m (average colour reduction p.m of 19%) Excellent decolourisation of up to 54% was achieved during the trial with Bulab 5154 dosed at 120 to 150 mg/l. If an average percentage colour removal is calculated for the period pre-, during and post-trial a higher percentage average colour removal of approximately 43% was observed during the period of Bulab 5154 addition compared with colour reduction around 34% pre- and 19% post-trial (see Figure 4) % Colour Reduced Pre-Bulab 5154 Addition Bulab 5154 Addition Post Bulab 5154 Addition 10 0 Figure 4. Decolourisation achieved before, during and after the plant trial with Bulab 5154 at Refinery A 462

7 During the plant trial there were a number of operational problems resulting in unscheduled stops and disrupted production which ultimately resulted in the trial being terminated. Since the trial was run towards the end of the season, a decision was taken to reschedule the trial for the new season. Further plant trials will be run to verify and confirm the results and establish the impact on the final refined sugar colour, since investigations thus far have only considered the colour reduction impact on the refinery liquors. Conclusions Laboratory investigations indicated that the cationic polymer precipitants were able to reduce the refinery melt colour by 30 to 57%. The product was effective in liquor from refineries employing different methods of refining; that is Refinery A which used a carbonatation followed by sulphitation process and Refinery B employing phosphatation followed by ion exchange. The addition of 200 mg/l Bulab 5154 to the carbonated liquor from Refinery A improved the decolourisation by 56%. The optimal product addition for Refinery B was found to be 150 mg/l Bulab Refinery B liquor colour was improved by up to 55% at this addition rate. A short plant trial has indicated that similar colour improvements of up to 54% could be seen when Bulab 5154 was applied at a dosage between 120 and 150 mg/l. The use of the colour precipitants may reduce chemical costs due to enhanced colour removal performance, thereby reducing the amount of refining chemicals required in the sulphitation or phosphatation phase. Previous investigations have suggested that colour precipitants such as polyamines have effectively replaced approximately 50% of the chemicals used during melt sulphitation (Moodley, 1993). Full scale plant trials are required to verify savings in the chemical spend. Alternatively, colour targets may be effectively reached during periods of high colour loading. Acknowledgements The authors would like to thank the technicians in the Innovation Division for their commitment, dedication and support in the evaluation of the new polymer colour precipitant and those who participated in the review of this paper. REFERENCES Anon (2009). Method 3.4 Juices: ICUMSA 420 colour. SASTA Laboratory Manual, including the Official Methods. Published by the South African Sugar Technologists Association, Mount Edgecombe, South Africa, 4pp. Bennett MC, Gardiner FJ, Abram JC and Rundell JT (1971). The Talofloc decolourisation process, Proc Int Soc Sug Cane Technol 14: Moodley M (1993). The application of cationic flocculants as decolourising agents in the sugar industry. Proc S Afr Sug Technol Ass 67: