WO2000066490A1

WO2000066490A1 - Regeneration of activated carbon

Info

Publication number: WO2000066490A1
Application number: PCT/NL2000/000284
Authority: WO
Inventors: Eduardus Gerardus Johannes Staal; Marcellinus Theodorus Carolus Scholten
Original assignee: Norit Nederland B.V.
Priority date: 1999-05-03
Filing date: 2000-05-02
Publication date: 2000-11-09
Also published as: AU4438000A; ZA200110400B

Abstract

The present invention relates to a process for regenerating activated carbon that has been used in decolourising clarified cane sugar syrup, wherein activated carbon is treated with an aqueous solution of a strong base, and neutralised. The invention further relates to the use of regenerated activated carbon obtained in said process for decolourising cane sugar.

Description

Title: Regeneration of activated carbon

The invention relates to a process for regenerating activated carbon, in particular of activated carbon used for decolourisation of raw cane sugar. The invention further relates to a process for decolourising raw cane sugar using activated carbon regenerated in said process. In the industrial manufacture of cane sugar, a raw sugar product is usually first prepared. This product is subsequently refined into final products. Raw sugar is traded on the commodity futures markets. White sugar products have traditionally been produced at sugar factories in the tropics for household and industrial use. Cane used for producing the raw sugar product, is usually first washed and milled, before it is decolourised. The juice, which is obtained after decolourisation, is subjected to evaporation and crystallization to obtain the raw sugar product.

Powdered activated carbon may be used for additional decolourisation before the crystallisation. Further, in a raw sugar refining process, raw sugar, e.g. obtained as described above, is melted, and the obtained melt is clarified and decolorized, before it is crystallised into refined white sugar. Molasses is a by-product obtained during refining. Brown (painted) sugar and confectionery syrups are products obtained by further processing the sugar and the by-products. Particularly, activated carbons are extremely efficient adsorbents for the decolourisation and purification of these sucroses or sucrose based liquors. Powdered or granular activated carbons are often used for these decolourisation processes.

The present invention relates to the regeneration of activated carbon used in any of these types of decolourisation processes, more in particular to decolourising dissolved, washed raw cane sugar.

In order to be able to more closely regulate the pH of clarified and filtered cane sugar syrup, and other sucrose based products, it is known to use dead-burned magnesite in amounts of 5-10% together with the activated carbon. The purpose of this is to prevent the sucrose from hydrolysis, resulting in inverted sugar. Apart from sucrose loss, downstream problems may occur when inverted sugar products are present in the decolourised syrup. A process involving the use of dead-burned magnesite is described in the US patent 2,822,304. In accordance with the invention, use is made of a similar mixture of magnesite and activated carbon for decolourising clarified cane sugar.

After a certain period of time, the activated carbon in operation is not providing enough adsorption capacity any longer and must be replaced. In the traditional granular carbon operations, the exhausted carbon (spent carbon) is regenerated. This is usually done by thermal regeneration in dedicated reactivation kilns at typically 950°C in the presence of steam.

The present invention aims at providing a novel manner of regenerating activated carbon used in the decolourisation of clarified cane sugar syrup. Desirably, the objective process leads to a regenerated activated carbon which has substantially the same, or at least sufficient, capacity for decolourising clarified cane sugar syrup when compared to fresh (virgin) activated carbon. Further, the objective process should be flexible, so that it can be adjusted to the different circumstances that exist in different cane sugar processing industries. In addition, it is desired that the process leads to an activated carbon which can be regenerated many more times.

Surprisingly, it has now been found that the above objects are reached by the provision of a process for regenerating activated carbon which involves an alkaline treatment and a neutralisation step. Accordingly, the invention relates to a process for regenerating activated carbon that has been used in decolourising a raw sugar cane product, wherein activated carbon is treated with an aqueous solution of a strong base, and neutralised.

The present process has been found to lead to an activated carbon which is highly active in the decolourisation of clarified cane sugar. Also, it is a flexible process, which can conveniently be incorporated in any cane sugar processing industry. Furthermore, the present regeneration process may be carried out in granular carbon columns, resulting in a strong simplification of the process design, as reactivation kilns, transport and storage facilities can be omitted from the design. Accordingly, hydraulic transport to and from reactivation ovens and possible intermittent storage, which may give rise to loss or abrasion of activated carbon, is no longer necessary.

The activated carbon that is regenerated in a process according to the invention may be any type of carbon that can be used in the decolourisation of clarified cane sugar. However, the higher the macroporosity of the carbon, the better the re gener ability according to the present invention. Very suitable are activated carbons produced by the phosphoric acid activation process. Suitable types include granular activated carbon having a particle size (D50) between 0.9 and 1.5 mm, preferably of about 1.2 mm. The D50 is defined as the value at which 50 wt.% of the activated carbon does not go through a sieve having a sieve size equal to said value. The particle size distribution will generally be between 0.42 and 2.4 mm, preferably between 0.8 and 1.7 mm. The upper and lower limits of the range of the particle size distribution reflect the situation at which less than 10 wt.% of the activated carbon is outside the indicated range (5 wt.% below and 5 wt.% above). The total pore volume will usually be between 1.5 and 2.0 mL/gram, indicating that the pore diameters are smaller than 15 μm, or that the pore radii are smaller than 7500 nm.

As has been mentioned above, it is preferred that the activated carbon bed comprises magnesite. The magnesite (or magnesium oxide) may be of any known type, such as caustic magnesite or dead-burned magnesite. Preferably, dead-burned magnesite is used. This is a type of magnesite composed of dense hard grains or granules which do not disintegrate or abrade under the severest handling conditions. It is stable at elevated temperatures in the presence of steam, air and combustion gases. Also, it maintains its range of limited solubility. Dead-burned magnesite is generally prepared by calcining magnesite at 1540-1845°C, preferably at about 1650°C. It can also be prepared by similarly heating chemically precipitated magnesium hydroxide. It is preferred that the dead-burned magnesite has a calcium oxide content below 2% and an iron oxide content also below 2%. The iron content is kept at a minimum to prevent contamination, as acidic liquors leach out of the iron. On the other hand, if too much calcium oxide is present, the alkalinity of the liquors will be increased.

The magnesite is preferably added to the activated carbon in an amount of 5-20 wt.%, more preferably 10-15 wt.%, based on the activated carbon. It has been found that these amounts of magnesite lead to an activated carbon having a very good activity in the decolourisation of cane sugar. The magnesite and activated carbon may have the same particle size, but it has frequently been found advantageous to use dead-burned magnesite having a smaller particle size than the activated carbon, in order to give equal settling velocities. A preferred particle size for the magnesite lies within the range of 0.25-1.0, more preferably of 0.3-0.9 mm. It has been found that the use of magnesite having a particle size within this range leads to an optimal control of the pH during the decolourisation.

The admixing of the activated carbon and magnesite may be carried out in any conventional way, for instance by tumbling the two materials together or by charging the materials simultaneously to a vessel, or a bed in a column to be employed for the decolourisation process.

Prior to carrying out the present process for regeneration, it is desired to recover all cane sugar present in the column, in which the activated carbon is employed for decolourising the sugar. This, which is often referred to as "sweetening off, or " desweetening", may be accomplished in any conventional manner.

In case a substantial decrease in adsorptive capacity is noticed after several regenerations in accordance with the invention, and this decrease is identified as being caused by accumulation of minerals such as calcium phosphate, an occasional acid flush can be carried out. This flush can be done prior to the regeneration step. After this step, the activated carbon will usually be washed with water.

Subsequently, the activated carbon is subjected to an alkaline treatment, such as a treatment with an aqueous solution of a strong base. In principle, any strong base that is acceptable for use in the processing of food products may be used. A suitable example is sodium hydroxide. Preferably, the solution comprises 2 wt.%, based on the weight of the solution, of sodium hydroxide. In a preferred embodiment, the alkaline treatment is carried out by leading the alkaline solution in countercurrent through a bed filled with the activated carbon. The amount of alkaline solution used in this manner is preferably about 4 bed volumes.

After the alkaline treatment, it is preferred that another wash step is carried out. This may be done by leading about 2 bed volumes of water over the bed of activated carbon, in the same direction as the alkaline solution. Advantageously, this wash step removes excess alkaline solution and contaminants and provides a (partial) neutralization.

Finally, the activated carbon is neutralised, preferably to a pH of

6-9, more preferably of 7-8. This is a very important step of the present process. Should the regenerated activated carbon eventually have a too high pH, then the decolourisation of cane sugar does not proceed in an optimal manner. A pH which is too low leads to inversion and, consequently, to a loss of sugar. Preferably, an acid is chosen which buffers at a pH around 7. Although it is possible to use any type of acidic solution for the neutralisation, such as a phosphoric acid solution comprising 0.3 wt.% phosphoric acid, it is preferred that an aqueous solution of a weak acid is used. Examples of suitable weak acids in this regard are citric acid, formic acid, acetic acid and propanoic acid. Preferably, an aqueous solution containing about 0.8 wt.% citric acid is used, as this acid substantially does not react with any calcium present. The neutralisation may be carried out by leading an aqueous solution containing the chosen acid over the bed of activated carbon. This can be done in any current direction (either countercurrent or not). The pH is preferably measured on the effluent side of the bed. It has been found that about 1-2 bed volumes are usually sufficient to reach a pH of about 7-8 after a subsequent washing step with water.

The regenerated activated carbon may be washed with water after the neutralisation step, in order to remove traces of acid. In a preferred embodiment, this wash step is used to adjust the pH to substantially neutral.

The thus obtained regenerated activated carbon may be used as such in the decolourisation of clarified cane sugar. It will be understood that the invention also pertains to the use of said chemically regenerated activated carbon in a decolourisation process of clarified cane sugar syrup.

The invention further relates to a process for decolourisation of clarified cane sugar in which the regeneration, as described above, of activated carbon is implemented in a very efficient manner. This process involves a setup of a multiple column system in which a bed of activated carbon is present. A multiple column system is designed in such a way that a continuos decolourisation of clarified cane sugar syrup is established without interruption of the process. Columns can be switched into service, while the columns that need to be regenerated can be taken out of service, are desweetened and regenerated at the same time. In view of the fact that the activated carbon cannot be regenerated endlessly and fresh activated carbon will have to be introduced into the decolourisation process at some point, it is preferred that the decolourisation makes use of columns operating in series, allowing for maximum adsorption performance and efficient regeneration as described above. This type of preferred serial operation works with ead' and lag' columns. The feed for the lead column is the clarified cane sugar syrup and the feed for the lag column is the effluent of the lead column. In the lag column, the cane extract juice will be further decolourised until a desired colour is achieved, which is often below 50% of the original value. A value below 300 ICUMSA is often regarded sufficient for further processing of the decolourised sugar product in most cane sugar producing industries, to obtain the quality of granulated, refined sugar after crystallisation.

In the context of the invention, colour is expressed in ICUMSA and can be measured as described in "Sugar Analysis", F. Schneider (Ed.), ICUMSA, Peterborough, 1979, pp.125-128, after membrane filtration at a wavelength of 420 nm and a pH of 7.

It is expected that, depending on the process conditions during decolourisation, in general the activated carbon used in this lead column can be regenerated at least 10 times before fresh activated carbon will have to be introduced. After refilling with fresh activated carbon, the former lag columns will operate as lead columns, and the refilled columns wil operate as lag column.

It is expected that the activated carbon used in the second step will have a longer lifetime before regeneration is required than the activated carbon used in the first step. Therefore, it is preferred that, when the activated carbon of both columns of the first step can no longer be regenerated and fresh activated carbon must be introduced, the columns of the second step are employed for the first rough decolourisation of the first step. The columns which were previously used in the first step are then reloaded with fresh activated carbon and deployed for the second, finer decolourisation. It has been found that this process design allows for a very efficient use of activated carbon and also allows for a very high yield of decolourised sugar product per amount of activated carbon and per time unit.

The invention will now be elucidated by the following, non- restrictive examples. The tests of the examples have been performed in an apparatus comprising 4 glass columns in series. Each column had an inner diameter of 6.0 cm and a height of 100 cm. The columns were water jacketed in order to be able to maintain a test temperature of 75°C. This temperature was controlled with a water bath (JUBALO, model WM-12) adjusted at 75°C. A peristaltic pump (WATSON MARLO, model 505U with 501RL pump head) equipped with Marprene pump tubing was used to pump all liquids through the carbon bed. The flow during sugar syrup filtration was controlled by means of a magnetic flow controller (YOKOGAWA, model AE102DG) and an industrial controller (PHILIPS, model KS 40) which responds to the conductivity of the sugar syrup. The flow during the other steps was manually controlled. The apparatus is also capable of working in countercurrent.

Example 1

Each column has been filled with 450 gram activated carbon which has been de-aerated by boiling in water for 10 minutes. To each column 55 gram Magnesite (M.A.F. Magnesite B.V. Vlaardingen, type (Dead Burned) NedMag 99, sieve fraction 0.43-0.84 mm) has been added. Each column has been carefully backwashed.

The fresh carbon has been pre-treated with a NaOH solution in water of 2.0 w/w% (1020 gram NaOH technical grade per 50 liter water) which has been pumped through the carbon beds at a velocity of 1.33 bed volumes per hour during 3 hours. The carbon has been washed by means of 2 bed volumes of water for 1 hour. The neutralisation of the carbon has been performed with a 0.3 w/w% solution of phosphoric acid in water (90 gram of 85% HsPO solution per 25 liter water). During this step the pH of the effluent drops from above 12 to around 11.

Subsequently, the carbon has been washed with water for 2 bed volumes in 1 hour. During this step the pH drops to 7.

Sugar syrup (approx. 65 Brix, pH approx.7) with an average colour of 550 ICUMSA has been filtered through the carbon bed at a contact time of 3 hours for a period of 8 days. In this period a total of 540 litre sugar syrup has been filtered. During this period the sugar syrup was decolourised to an average colour of 205 ICUMSA (average decolourisation 63%).

The loaded carbon which remained in the columns has been sweetened off with 2 bed volumes of demi water in 1 hour (flow 2.0 bed volumes per hour). A NaOH solution in water of 2.0 w/w% (1020 gram NaOH technical grade per 50 litre water) has been pumped through the carbon beds at a velocity of 1.33 bed volumes per hour during 3 hours.

The carbon has been washed by means of 2 bed volumes of water for 1 hour. The neutralisation of the carbon has been performed with a 0.3 w/w% solution of phosphoric acid in water (90 gram of 85% H₃PO₄ solution per

25 litre water). During this step the pH of the effluent drops from above 12 to around 11. Subsequently, the carbon has been washed with water for 2 bed volumes in 1 hour. During this step the pH drops to 6.7.

The regenerated carbon has been used to decolourise sugar syrup again. A syrup having an average colour of 505 ICUMSA could be passed through the carbon bed for 9 days. In this period 575 liter sugar syrup has been pumped through the carbon bed. The average effluent colour was 238 ICUMSA (average decolourisation 53%). The first sugar syrup effluent had a pH of 6.7, 6.2, 5.9 and 6.0 respectively for column 1, 2, 3 and 4.

The above described has been repeated twice resulting in an average decolourisation of 46% during a period of 10 days and 44% during a period of 7 days respectively for cycle 3 and 4. The pH of the first sugar syrup effluent of cycle 3 was 6.7, 6.2, 5.9 and 6.0 for resp. column 1, 2, 3 and 4. The pH of the first sugar syrup effluent of cycle 4 was 6.8, 5.9, 5.6 and 5.4 for resp. column 1, 2, 3 and 4.

Example 2:

Carbon loaded and regenerated for 7 times has been sweetened off.

The carbon has been washed with 1 w/w% HC1 (ON kg 36 w/w% HC1 per 25 liter water) to remove in-organic contaminants basically Calcium Phosphate. A ΝaOH solution in water of 2.0 w/w% (1020 gram ΝaOH technical grade per 50 litre water) has been pumped through the carbon beds at a velocity of 1.33 bed volumes per hour during 3 hours.

The carbon has been washed by means of 2 bed volumes of water for 1 hour. The neutralisation of the carbon has been performed with a 0.8 w/w% solution of citric acid in water (215 gram of 99.5% C6HsO .H2O per 25 liter water). During this step the pH of the effluent drops from above 12 to around

11. Subsequently, the carbon has been washed with water for 2 bed volumes in

1 hour. During this step the pH drops to 7. The regenerated carbon has been loaded during 10 days with

670 litre sugar syrup having an average influent colour of 414 ICUMSA. The average effluent colour after carbon filtration was 202 ICUMSA (average decolourisation 51%).

Claims

1. A process for regenerating activated carbon used for decolourising a raw cane sugar product, wherein said activated carbon is treated with an aqueous solution of a strong base, and neutralised.

2. A process according to claim 1, wherein the mixture is washed before treatment with the aqueous solution of a strong base.

3. A process according to claim 1 or 2, wherein the mixture is washed after treatment with the aqueous solution of a strong base.

4. A process according to any of the preceding claims, wherein the activated carbon is granular activated carbon having a particle size(Dso) of 0.9 to 1.5 mm and a particle size distribution of 0.42 to 2.4 mm.

5. A process according to claim 4, wherein the activated carbon has a total pore volume of 1.5 to 2.0 ml/gram.

6. A process according to any of the preceding claims, wherein the activated carbon comprises magnesite in an amount ranging from 5 to 20 wt.%, preferably 10 to 15 wt.%, based on the weight of activated carbon.

7. A process according to any of the preceding claims, wherein the strong base is sodium hydroxide.

8. A process according to any of the preceding claims, wherein the aqueous solution of sodium hydroxide comprises between 1 and 10 preferably below 5 wt.% of sodium hydroxide, based on the weight of the solution.

9. A process according to any of the preceding claims, wherein the neutralisation is carried out using an acid, chosen from the group of citric acid, formic acid, acetic acid, propanoic acid and phosphoric acid.

10. The use of regenerated activated carbon obtained in a process according to any of the claims 1-9 in the decolourisation of clarified juice of cane sugar.

11. A process for decolourising cane sugar, wherein in a first step raw cane sugar product is decolourised using activated carbon and wherein the effluent of the first step is fed to a second step, wherein it is further decolourised using activated carbon to a value below 50% of the original value, and wherein the activated carbon is regenerated in a process according to any of the claims 1-9.

12. A process according to claim 11, wherein the first and second step each comprise the use of at least two decolourisation columns, allowing for simultaneous decolourisation of cane sugar and regeneration of activated carbon.