WO2024246498A1

WO2024246498A1 - A method for the reduction of sugar content in a comestible liquid

Info

Publication number: WO2024246498A1
Application number: PCT/GB2024/051344
Authority: WO
Inventors: Darren Reed; Samson Olufemi Abiodun Junior AKANBI
Original assignee: Madison Filter 981 Ltd
Current assignee: Madison Filter 981 Ltd
Priority date: 2023-05-26
Filing date: 2024-05-24
Publication date: 2024-12-05
Anticipated expiration: 2025-11-26
Also published as: GB202307907D0; GB2630378A

Abstract

21 ABSTRACT A METHOD FOR THE REDUCTION OF SUGAR CONTENT IN FRUIT JUICE A method for the reduction of sugar content in comestible liquid, comprising the steps of subjecting at least a portion of a comestible liquid containing sugar to at least one 5 nanofiltration step to produce a nanofiltered retentate and a nanofiltered permeate, then subsequently subjecting at least a portion of the nanofiltered retentate to at least one ultrafiltration step to produce an ultrafiltered retentate and an ultrafiltered permeate. FIG 1 10

Description

A METHOD FOR THE REDUCTION OF SUGAR CONTENT IN A COMESTIBEE EIQUID

Technical Field of the Invention

The present invention relates generally to the field of processing of comestible liquids, particularly for the at least partial modification of their composition, by adjustment of the content of one of their constituents and particularly to a method for the controlled reduction of the sugar content of juice.

Background to the Invention

In recent times, there has been more of a focus on and a consumer drive toward lower sugar content, lower calorie content and/or lower alcohol in foods and beverages.

In seeking to lower sugar content in comestible liquids used in or as foods and beverages, care should be taken to minimise any negative impacts on the organoleptic properties. This is particularly important for comestible liquids which rely on the organoleptic properties, of which wine is an example.

It would therefore be an advance in the art to reduce the sugar content of a comestible liquid whilst minimising any negative impacts on the organoleptic properties.

Existing design solutions currently on the market do not satisfy the above requirements and the new design below solves all of the current issues.

Summary of the Invention

According to one aspect of the invention, there is provided a method for the reduction of sugar content in a comestible liquid, comprising the steps of subjecting at least a portion of a comestible liquid containing sugar to at least one nanofiltration step to produce a nanofiltered retentate and a nanofiltered permeate, then subsequently subjecting at least a portion of the nanofiltered retentate to at least one ultrafiltration step to produce an ultrafiltered retentate and an ultrafiltered permeate.

The method may further include the step of blending of at least a portion of the ultrafiltered retentate with at least a portion of the nanofiltered permeate to create the sugar reduced liquid. The method may be applied to a comestible liquid containing sugar particularly fruit juice, which includes fruit must, which is used to make wine. The method provided will preferably produce a concentrated juice retentate after nanofitration with a primarily water, nanofiltered permeate, and a low volume concentrated juice retentate and a sugar water permeate following ultrafiltration. The removal of the sugar water permeate in the ultrafiltration step will reduce the sugar content in the ultrafiltered retentate. The ultrafiltered retentate can be mixed with the nanofiltered permeate to further adjust the sugar content (by adjusting the volume).

The main disadvantage of nanofiltration, as with all membrane filter technology, is the cost and maintenance of the nano filtration membranes. These membranes are an expensive portion of the process. Repairs and replacement of membranes depend on total dissolved solids, flow rate, and feed components. It is conventional practice for comestible liquids to undergo separation using membranes with successively smaller pore sizes as this results in each membrane removing only a portion of the contents, with the next smaller pore size membrane removing the next cut of sizes.

Use of a nanofiltration membrane before an ultrafiltration membrane is contrary to convention knowledge, primarily because the pores of the nanofiltration membrane will ‘clog’ faster if located before the ultrafiltration step as the nanofiltration membrane will retain larger contents. The nanofiltration membrane would then require cleaning or flushing more often and may be at greater risk of damage if provided before an ultrafiltration membrane, due to the greater load placed on the nanofiltration membrane due to removing a greater amount of permeate.

In this document, the term ‘ultrafiltration’ is used to designate a separation process using a membrane with a pore size of approximately 0.01 micron.

In this document, the term ‘nanofiltration’ is used to designate a separation process using a membrane with a pore size of approximately 0.001 micron.

Degrees Brix (symbol °Bx) is a measure of the dissolved solids in a liquid and is commonly used to measure dissolved sugar content of an aqueous solution. One degree Brix is 1 gram of sucrose in 100 grams of solution and represents the strength of the solution as percentage by mass. The °Bx is traditionally used in the wine, sugar, carbonated beverage, fruit juice, fresh produce, maple syrup and honey industries.

Comparable scales for indicating sucrose content are the Plato scale (°P), which is widely used by the brewing industry; the Oechsle scale used in German and Swiss wine making industries, amongst others; and the Balling scale.

Depending upon the particular feed material to be subjected to the method, the operating temperature may preferably remain at or below 25 °C at all times throughout the method. However, there may be times when a feed material can be processed at a higher operating temperature as the effectiveness of the filtration will preferably increase at higher temperatures, that is Flux at the same VCF (Volumetric Concentration Factor) is higher at higher temperatures than at lower temperatures.

One or more heat exchangers may be utilised in one or both of the nanofiltration and ultrafiltration steps to control the operating temperature of at least the retentate in each step. One or more heat exchangers may be utilised in the nanofiltration step to control the temperature of the nanofiltered permeate to be added back to the ultrafiltration retentate in the mixing step.

The particular membranes selected for each of the nanofiltration and ultrafiltration steps may preferably be selected according to the feed material type.

The method may further comprise one or more cleaning steps before use or between uses. An appropriate cleaning solution will be used, primarily according to the membrane type as each membrane will normally have a pH limit and/or other limits which should be observed in order to minimise or prevent membrane damage or degradation.

The cleaning solution may be enzyme based, caustic or alkaline for some membranes and acidic for other membranes. An example of a caustic or alkaline solution which could be used is the Ultrasil range made Ecolab® (who also make acidic cleaning solutions). In a preferred form, a caustic or alkaline solution may be used for approximately 30 mins. A caustic or alkaline solution of approximately pH 9.5 may be used at a temperature above the operating temperature, for example approximately 30°C to 45°C. Where an acidic cleaning solution is used, an acidic solution of approximately pH 2.5 may be used at a temperature above the operating temperature, for example approximately 40-55°C. In a preferred form, an acidic solution may be used for approximately 30 mins. Nitric acid is an example of a possible acidic solution.

The method may further comprise one or more flushing steps, before use and/or between uses. The or each flushing step may utilise clean water, preferably clean water. The water may be fresh water, free of chemicals such as Chlorine.

In one form, the method may follow a flush step-clean step-flush step regime or pattern. Alternatively, a clean step-flush step regime may be used. In some cases, a flush step only regime may be used, particularly between uses.

Any one or more of the regimes outlined above may act to regenerate the membrane so that membrane efficiency is maintained between uses.

The method preferably includes the step of subjecting at least a portion of a comestible liquid containing sugar to nanofiltration to produce a nanofiltered retentate and a nanofiltered permeate.

The nanofiltered permeate is preferably only water with very few components removed from the feed. Removal of water from the feed acts to decrease the volume of the nanofiltered retentate which is then subjected to the ultrafiltration step. This preferably concentrates the desired components in the nanofiltered retentate.

The operating parameters of the nanofiltration step preferably depend on the feed material. The important operating parameter in the nano filtration step is preferably pressure. For most comestible liquids, a pressure of between 35 to 60 bar is used. The optimum pressure for different comestible liquids differs though. For example, a pressure of 45 to 52 bar may be preferred or white must and a higher pressure of 48 to 55 bar may be preferred for rose must.

A preferred volumetric concentration factor for the nanofiltration stage may be between 1.50 and 3.00 VCF or higher depending on the starting °Brix value of the comestible liquid. A volumetric concentration factor varying between 1.50 and 3.50 can be used for the ultrafiltration stage depending on the °Brix value of the nanofiltered retentate. Volumetric concentration factor (VCF) is calculated as follows: VCF = Initial total volume

Product volume at time x

Other important parameters may be temperature and crossflow velocity.

The operating parameters are preferably manipulated to achieve an optimum average flux, so that a desired output volume can be achieved in a given time.

The method preferably includes the step of subsequently subjecting at least a portion of the nanofiltered retentate to ultrafiltration to produce an ultrafiltered retentate and an ultrafiltered permeate.

The pressure utilised in the nanofiltration step is important to minimise or prevent damage to the nanofiltration membrane.

Given that the nanofiltration step occurs before the ultrafiltration step, additional care should be taken with regard to the pressure used in the ultrafiltration step to minimise or prevent damage to the ultrafiltration membrane. The operating pressure in the ultrafiltration step will preferably be lower than that used in the nanofiltration step.

A pressure range of 7 to 20 bar may be used in the at least one ultrafiltration step but the actual pressure used will depend on the feed liquid. The ultrafiltration of the white must may be completed with a pressure range of 7 - 8.5 bar, whilst the rose must may be done between 8.5 - 10 bar.

A preferred volumetric concentration factor for the ultrafiltration stage may be between 2.5 and 3.3 depending on the °Brix value of the nanofiltered retentate.

The process parameters, particularly processing pressure are preferably selected according to the feed type and the membrane used.

The method may further include the step of blending of at least a portion of the ultrafiltered retentate with at least a portion of the nanofiltered permeate to create the sugar reduced liquid.

The permeate fluxes, as the feed/retentate concentrates, may decrease within expected and normal concentration process. The temperature of the product to be concentrated may influences the permeate flows, which may be higher at higher initial temperatures.

The sugar reduction targets of 30% (or greater) can be met, obtaining values in the nanofiltered permeate of 0-2°Brix, and a concentrate with values of 22- 24°Brix, and in the ultrafiltered permeate with values of 20-22°Brix and a concentrate of 24- 27°Brix. Values will vary at targets above 30%.

The permeate obtained in nanofiltration may be completely colourless, unlike in the ultrafiltration stage, where a colouring of the permeate obtained may be observed. This is to be expected however, as UF membranes are more open, allowing more of the colour components to pass into the permeate.

The permeate obtained in ultrafiltration may be used as a secondary product, as the permeate obtained in ultrafiltration may have values of 20-22°Brix.

The membranes have a very good regeneration after chemical cleaning.

The regeneration of the membranes after flushing with water, preferably hot water may be effective, without chemical cleaning.

Detailed Description of the Invention

In order that the invention may be more clearly understood one or more embodiments thereof will now be described, by way of example only, with reference to the accompanying drawings, of which:

Figure 1 is a flow diagram of a method for the controlled reduction of the sugar content of juice according to an embodiment.

Figure 2 is a graphical representation of the MSR (Nanofiltration) Process Flow Diagram for Example 2.

Figure 3 is a graphical representation of the UF15 (Ultrafiltration) Process Data for Example 2.

Figure 4 is a graph showing Flux vs. Volumetric Concentration Factor across the AFC 30 (NF) membrane in Run 1 to 4 Example 2. Figure 5 is a graph showing Flux vs. Volumetric Concentration Factor. PU120 (UF) membrane in Run 1 to 4 Example 2.

Figure 6 is a process overview from the test completed on the rose must in Example 2.

Figure 7 shows the NF & UF Process Steps and Mass Balance - Run 1, White Must A in Example 2.

Figure 8 shows the NF & UF Process Steps and Mass Balance - Run 2, White Must B in Example 2.

Figure 9 shows the NF & UF Process Steps and Mass Balance - Run 3, Rose Must in Example 2.

Figure 10 shows the NF & UF Process Steps and Mass Balance - Run 4, Rose Must in Example 2.

A method for the reduction of sugar content in comestible liquid is provided.

As illustrated in Figure 1 with reference to orange juice as the feed liquid, the method may comprise the steps of subjecting the feed orange juice containing sugar to a nanofiltration step to produce a nanofiltered retentate which is concentrated orange juice, and a nanofiltered permeate which is basically water, then subsequently subjecting at least a portion of the nanofiltered retentate to an ultrafiltration step to produce an ultrafiltered retentate which is low volume concentrated orange juice and an ultrafiltered permeate which is sugary water.

The method may further include the step of blending of at least a portion of the ultrafiltered retentate (low volume concentrated orange juice) with at least a portion of the nanofiltered permeate (water) to create the sugar reduced liquid which retains the organoleptic qualities of the original feed orange juice with a sugar content which is 30-40% lower in sugar.

The removal of the sugar water permeate in the ultrafiltration step will reduce the sugar content in the ultrafiltered retentate. The ultrafiltered retentate can be mixed with the nanofiltered permeate to further adjust the sugar content (by adjusting the volume).

Depending upon the particular feed material to be subjected to the method, the operating temperature will preferably remain at or below 25 °C at all times.

The cleaning solution may be enzyme based, caustic or alkaline for some membranes and acidic for other membranes. An example of a caustic or alkaline solution which could be used is the Ultrasil range made Ecolab® (who also make acidic cleaning solutions). In a preferred form, a caustic or alkaline solution may be used for approximately 30 mins. A caustic or alkaline solution of approximately pH 9.5 may be used at a temperature above the operating temperature, for example approximately 30°C to 45°C.

Where an acidic cleaning solution is used, an acidic solution of approximately pH 2.5 may be used at a temperature above the operating temperature, for example approximately 40-55°C. In a preferred form, an acidic solution may be used for approximately 30 mins. Nitric acid is an example of a possible acidic solution.

The method may further comprise one or more flushing steps, before use and/or between uses. The or each flushing step may utilise clean water, preferably clean water. The water may be fresh water, free of chemicals such as Chlorine. In one form, the method may follow a flush step-clean step-flush step regime or pattern. Alternatively, a clean step-flush step regime may be used. In some cases, a flush step only regime may be used, particularly between uses.

The pressure utilised in the nanofiltration step is an important parameter to optimise to minimise or prevent damage to the nanofiltration membrane.

As the nanofiltration step occurs before the ultrafiltration step, additional care should be taken with regard to the pressure used in the ultrafiltration step to minimise or prevent damage to the ultrafiltration membrane. The operating pressure in the ultrafiltration step will preferably be lower than that used in the nanofiltration step.

The process parameters, particularly processing pressure, are preferably selected according to the feed type and the membranes used.

Undertaking the steps with nanofiltration prior to ultrafiltration can achieve a sugar reduction target of 30% or greater.

The permeate obtained in ultrafiltration may be used as a secondary product.

The membranes have a very good regeneration after chemical cleaning, however, the regeneration of the membranes after flushing with water, preferably hot water may be effective, without chemical cleaning.

Aspects of the invention may be further illustrated with reference to the following Examples: Example 1: Orange Juice

AFC30 membranes from PCI Membranes were used for the nanofiltration stage and installed a set into a 4ft B 1 module (B 1 Series Tubular Membrane Module from PCI Membranes). An acidic (pH 2.75, temperature 42 °C, 30 minutes) clean using nitric acid was performed to remove preservative from the membrane(s) for 30 minutes followed by a clean water flush. Next, a pressure scan was conducted to ascertain the best operating pressure for the trial (Table 1).

Based on the permeate flow results of the pressure scan, a ‘module in’ pressure of 40 bar was selected for the trial. The nanofiltration stage started by using 76 litres of orange juice. Juice analysis is reported in table 2.

This stage lasted approximately one hour and twenty minutes, in which the feed volume was reduced to 39 litres with 37 litres (30.5 litres plus circa 6.5 litres collected from the module shroud) permeate removed; equivalent to a volumetric concentration factor of 1.67 VCF. Permeate was collected in a 5-litre container (before transfer to a larger container) and timed to calculate flow rate. All permeate collected from the nanofiltration stage had 0 °Bx, with an average conductivity of 890 pS/cm.

The Bl module and lines were drained into relevant containers, the system was flushed with fresh water and then cleaned. The single tube tester was refitted and installed with PU120 membranes from PCI Membranes. The system was cleaned with alkaline solution (pH 9.5, temperature 35 °C, 30 minutes) and then flushed with clean water. The nanofiltration stage retentate (39 litres) was used as feed for the UF stage. Feed analysis is reported in table 3.

Stage 2 lasted approximately one hour and 5 minutes and resulted in the removal of circa 8 litres permeate. These 8 litres of ‘sweet, mildly orange coloured water’ are considered waste and are essentially the difference between the starting liquid and the final product. As seen in Figure 1, the stage 1 permeate and stage 2 retentate were combined to produce 68 litres of reduced sugar orange juice. Analysis in table 4.

Results and Discussion

The 2-stage system seen in Figure 1 provided excellent results from both stages. Table 5 below highlights the concentration of the orange juice as the permeate is removed.

The permeate in this stage appeared colourless with minimal smell and zero sugar content.

Table 5. Orange juice concentration with nanofiltration membrane AFC30

A further 6.5 litres was drained from the B 1 shroud and added to the 30.5 litres to make up 37 litres of ‘water’ removed from the orange juice with a final volumetric concentration factor (VCF) of 1.95, analysis in Table 6.

Table 6. Overall nanofiltration permeate analysis

In the second stage of the trial, the choice of UF membrane was crucial as it was imperative to select a membrane that would allow sugar and water, but little else to permeate through. Tighter membranes allowed reduced sugar into the permeate, which would result is less sugar reduction to the final product, and looser membranes would allow more undesirable particles into the permeate. As seen in the earlier ‘unsuccessful’ single tube tester trials, the PU120, ES625 and FP100 all allowed circa 10 °Bx sugar into the permeate. The ES404 and ES209 allowed circa 6-8 °Bx into the permeate. From this information it was decided that the PU120 membranes provided the best ‘middle ground’ especially considering there was minimal difference between conductivity of all permeates.

Table 7 shows the removal of 8 litres of ‘sugary water’ in stage 2. The starting volume of 39 litres was reduced to 31 and the sugar content kept constant at 20.5 °Bx. The brix analysis of the first permeate sample collected is inaccurate, whilst other analysis was relatively consistent.

Table 7. Concentrated juice clarification with ultrafiltration membrane PU120* in this table Cumulative Permeate Volume expressed in (L) is actually (mL).

As seen in the previous single tube tester trials the permeate brix was circa 85 % of the feed when using PU120 membranes. As previously stated, the nanofiltration permeate and ultrafiltration retentate were combined to form the ‘product’ . The sugar content was reduced from 11.5 to 8.5 °Bx, a reduction of circa 26 %, whilst the volume was reduced from 76 to 68 litres, a reduction of just 10 %.

Conclusion

Early ultrafiltration trials were unsuccessful at providing a permeate as the final product, as it did not resemble orange juice as we know and purchase it. A 2-stage system was adopted to remove ‘pure’ water, followed by ‘sugary’ water. The removal of the ‘sugary’ water was the key in reducing the overall sugar content of the juice, the ‘sugary’ water had a much higher brix than that of the original feed therefore reducing the sugar concentration of the orange juice. The overall sugar content was reduced by circa 26 % whereas the volume reduced by circa 10 %. This equates to having a ratio of 1:1.25 of retentate juice from the NF & UF to the NF permeate. With system optimisation, a sugar reduction of greater than 30 % could be achieved.

Example 2: Grape Must (White and Rose)

Two semi-industrial pilot units were used for this trial, the so-called MSR pilot unit for the nanofiltration (AFC30 membrane from PCI Membranes) stage and UF15 for the ultrafiltration (PU 120 from PCI Membranes) stage. Both the MSR and UF15 pilot units adopt a multi-stage with recirculation design concept, with each unit having a two-stage filtration configuration that involves the use of a high-pressure feed pump and a recirculation pump on each stage that helps to provide the required crossflow velocity whilst also compensating the pressure drop across each module. Each pilot unit is equipped with 6 off 12-ft B l module (3 module per stage, for a total of 15.6 m² membrane area) and 2 off 12-ft B 1 type heat exchanger (1 heat exchanger per stage)

All newly installed membranes were flushed, cleaned chemically and flushed once more with service water to remove any residual traces of preservative/cleaning chemicals. A clean water flux check was then completed in order to record the initial performance of the membranes being used.

An overview of the process flow diagram of the MSR unit and the process data from the UF15 plant are reported in Figures 2 and 3 which show MSR (Nanofiltration) Process Flow Diagram and the UF15 (Ultrafiltration) Process Data respectively.

To keep the temperature of the grape must below or equal to 25°C when processing the must with the pilot units, the use of a B 1 type heat exchanger with cooling fluid (glycol) was necessary at all times. In overall, the process fluid temperature has been kept below 25 °C at all times.

During this trial, two variety of white must, named: white must A and white must B, for the sake of this report, and a type of rose must have been tested. As the main objective of this trial is to evaluate the feasibility of reducing sugar from the grape must with membrane technology; thus, the procedures adopted with the filtration process involve an initial concentration of the grape must with a nanofiltration membrane up to a certain concentration factor to generate the so called: concentrate NF (CNF) and permeate NF (PNF), the resulting concentrate NF from the nanofiltration phase is then subjected to further clarification with the use an ultrafiltration membrane to generate two streams of products named: permeate UF (PUF) and retentate UF (RUF). A final grape must with reduced sugar content is then produced by blending both the permeate from the nanofiltration stage (PNF) and the retentate from the ultrafiltration stage (RUF). A summary of the process data across the length of both the nanofiltration and ultrafiltration tests is shown in Figures 4 and 5 which show Flux vs. Volumetric Concentration Factor. AFC30 (NF), Run 1 to 4 and Flux vs. Volumetric Concentration Factor. PU120 (UF), Run 1 to 4 respectively.

Both Figure 4 and Figure 5 show the variation of flux against the volumetric concentration factor achieved during the nanofiltration and ultrafiltration test respectively. With regards to the nanofiltration test; as highlighted in Figure 4, whilst the final volumetric concentration factor achieved across all 4 process runs varies between 1.52 and 1.64 VCF, process Run 1 with white must A shows the lowest average flux in relation to the concentration factor: circa 5.05 L/m²/hr at 1.52 VCF, with pressure ranging between 40 - 45 bar, and this is likely due to a combination of different factors: lower operating pressure and the type of must being processed. The highest average flux in relation to the concentration factor has been achieved from process Run 2 with white must B; about 9.29 L/m²/hr average flux resulted whilst operating with a slightly higher pressure (45 - 51 bar) at 1.62 VCF. Process Run 3 and 4 have been completed with the rose must and as expected, the type, colour and the constituents of the must will have impact on the resulting process flux and separation characteristic. Thus, to speed up the concentration of the rose must, the operating pressure for the nanofiltration phase was then increased further to help overcoming the osmotic pressure of the fluid. Process Run 3 with the rose must resulted in an average flux of 5.91 L/m²/hr at 1.57 VCF, with an operating pressure that varies between 48 - 53 bar. With similar rose must for Run 4, an average flux of 6.61 L/m²/hr resulted across the length of the concentration: 1.64 VCF, with an operating pressure that varies between 49 - 55 bar.

Similar interpretation as per the above can be applied to the case of the ultrafiltration tests, as shown in Figure 5. At similar crossflow velocity and temperature range for all four UF runs, the average flux from both white must A and B, resulted to be higher when compared to that of the rose must, even at a slightly lower operating pressure. The drop in flux for white must B at 2.02 VCF was due to a reduction in process temperature and lower crossflow velocity initially used across one of the two stages of the UF plant - this was later rectified by increasing the applied crossflow velocity. An average flux of 28.78 L/m²/hr at 3.3 VCF with an operating pressure range of 7.2 - 7.8 bar resulted from Run 1 with the white must A. 26.56 L/m²/hr average flux has been achieved from white must B, at 2.8 VCF and operating pressure range of 7.8 - 8.4 bar. At 2.5 VCF; an average flux of 11.13 L/m²/hr at an operating pressure range of 8.4 - 9.8 bar, and 8.34 L/m²/hr at 9 - 9.6 operating pressure range, have been achieved with the rose must for Run 3 and 4 respectively. In overall, under similar process operating conditions: pressure, temperature and crossflow velocity, the results achieved from phase two trial when processing the white must can be deemed comparable to those obtained from phase one trial.

Process overview from the test completed on the rose must and a summary of the mass balance across all process runs are shown in Figure 6.

Figure 7 shows the NF & UF Process Steps and Mass Balance - Run 1, White Must A.

Figure 8 shows the NF & UF Process Steps and Mass Balance - Run 2, White Must B.

Figure 9 shows the NF & UF Process Steps and Mass Balance - Run 3, Rose Must.

Figure 10 shows the NF & UF Process Steps and Mass Balance - Run 4, Rose Must.

As highlighted in Figure 6 and the tables in Figures 7 to 10, the implementation of two process steps, nanofiltration and ultrafiltration, would lead to achieving the requested sugar reduction from both white and rose grape must. On average, the sugar reduction across all processed must is in the region of 40%; whilst the volume reduction across each of these tests varies in relation to the selected volumetric concentration factor from the nanofiltration and ultrafiltration phase. The lowest and the highest volumetric reduction of 38% and 46% in relation to the starting must volume have been encountered for process Run 3 and 1 respectively. With regards to process Run 4, an actual volumetric reduction of 43.8% has been calculated; this takes into account the nanofiltration permeate volume used for the blending and the total volume processed with the ultrafiltration membrane. Some of the relevant data points in terms of Brix, VCF, blend ratio and volume are highlighted in green and red in the tables shown in Figures 7 to 10.

During this trial, a solution of caustic and dish washing liquid has been used for the cleaning in place (CIP) when required. For the nanofiltration membrane, the pH of the CIP solution was always kept below 9.5 - this was due to the maximum pH limitation of the NF membrane. However, the ultrafiltration membrane was mostly cleaned with a CIP solution having a pH range of 11 - 11.5. Following the initial CIP that was completed on the newly installed membranes, both the MSR and UF15 units were subsequently cleaned chemically at the end of process Run 2; in between these two runs, both plants were only flushed with cold and warm water. No CIP has been completed either between Run 3 and 4, as the units were mainly flushed with cold water only in between the two runs.

However, based on the performance of the tested membranes with both types of must, it is possible to deduce that the white must is less susceptible to fouling and hence, will require less frequent chemical clean. On the other hand; due to the nature of the rose must, a chemical clean will be required following every single process runs to help regenerate the membrane performance over time.

The results from this trial can used as further evidence to confirm that the implementation of a crossflow tubular membrane system to achieve the objective of sugar reduction from the grape must is feasible. Combining both nanofiltration and ultrafiltration processes as highlighted previously in this report, has proven to be suitable to achieve a sugar reduction of circa 40% across the length of all tested products. The blended volume from the nanofiltration permeate and the ultrafiltration retentate will result in approximately 38% to 46% less in relation to the starting volume - this of course, is dependent on the starting Brix of the must being processed, the type of must and the volumetric concentration factor across the NF and UF stages. Thus, it can be stated that the typical volumetric concentration factor for the nanofiltration stage is between 1.52 and 1.64 VCF, whilst a volumetric concentration factor varying between 2.5 and 3.3 can be used for the ultrafiltration stage.

As seen from the data highlighted in this trial; on average with the nanofiltration membrane, the white must has a higher flux and this correspond to about 1.4-fold to that of the rose must. Whereas, for the ultrafiltration stage, the white must exhibit an average flux of circa 2.84-fold to that of the rose must. Thus, the recommendation is to process the white must first before the rose must. In addition to this; for the nanofiltration stage, it is recommended to process the white must at a pressure range of 45 - 50 bar, whilst the rose could be processed at 50 - 55 bar pressure range. The ultrafiltration of the white must can be completed with a pressure range of 7 - 8.5 bar, whilst the rose must can be done between 8.5 - 10 bar. The requirement for a CIP after processing the white must could be based on a CIP per two batch operation, with sufficient hot water flush in between the two batches. However, the recommendation is to complete a CIP following every single batch of the rose must. All CIP should be completed with a solution that has a temperature range of 45 - 50°C, whilst also complying with the maximum acceptable pH for the membrane type being used.

For the full-scale plant, it is important to double check the mass balance data of both NF and UF stages before the completion of each batch and the blending of the final product. To keep the process temperature below or equal to 25°C at all times, the use of heat exchangers is preferred during both filtration steps.

The one or more embodiments are described above by way of example only.

Many variations are possible without departing from the scope of protection afforded by the appended claims.

Claims

1. A method for the reduction of sugar content in comestible liquid, comprising the steps of subjecting at least a portion of a comestible liquid containing sugar to at least one nanofiltration step to produce a nanofiltered retentate and a nanofiltered permeate, then subsequently subjecting at least a portion of the nanofiltered retentate to at least one ultrafiltration step to produce an ultrafiltered retentate and an ultrafiltered permeate.

2. A method as claimed in claim 1 further including the step of blending of at least a portion of the ultrafiltered retentate with at least a portion of the nanofiltered permeate to create the sugar reduced liquid.

3. A method as claimed in claim 2 wherein one or more heat exchangers are utilised to control the temperature of the nanofiltered permeate to be added back to the ultrafiltration retentate in the mixing step.

4. A method as claimed in any one of the preceding claims applied to a fruit juice.

5. A method as claimed in any one of the preceding claims wherein the at least one ultrafiltration step is undertaken using a membrane with a pore size of approximately 0.01 micron.

6. A method as claimed in any one of the preceding claims wherein the at least one nanofiltration step is undertaken using a membrane with a pore size of approximately 0.001 micron.

7. A method as claimed in any one of the preceding claims wherein the operating temperature remains at or below 25 °C at all times.

8. A method as claimed in any one of the preceding claims wherein one or more heat exchangers are utilised in one or both of the at least one nanofiltration step and at least one ultrafiltration step to control the operating temperature of at least the retentate in each step.

9. A method as claimed in any one of the preceding claims further comprising one or more cleaning steps before use or between uses.

10. A method as claimed in any one of the preceding claims further comprising one or more flushing steps, before use and/or between uses.

11. A method as claimed in claim 10 wherein the one or more flushing steps utilises heated water.

12. A method as claimed in claim 10 or claim 11 wherein the method follows a flush step-clean step-flush step regime.

13. A method as claimed in claim 10 or claim 11 wherein the method follows a clean step-flush step regime.

14. A method as claimed in any one of the preceding claims wherein, in the nanofiltration step, a pressure of between 35 to 60 bar is used.

15. A method as claimed in any one of the preceding claims wherein a volumetric concentration factor between 1.50 and 3.00 is used for the nanofiltration stage.

16. A method as claimed in any one of the preceding claims wherein a volumetric concentration factor between 1.50 and 3.50 is used for the ultrafiltration stage.

17. A method as claimed in any one of the preceding claims wherein an operating pressure in the at least one ultrafiltration step is lower than that used in the at least one nanofiltration step.

18. A method as claimed in any one of the preceding claims wherein a pressure range of 7 to 20 bar is used in the at least one ultrafiltration step.