Adsorbents for Treating Contaminated Liquids

Abstract

There is described a method of making an adsorbent material comprising mixing first particulate material with a second material, homogenising the mixture of the first and second materials, incorporating an impregnating or coating material capable of carbonisation, and carbonising the mixture. Also described are adsorbent materials manufactured according to said method and the use of such adsorbent materials in the treatment of a contaminated liquid. Further described is a method of removing contaminants from a quantity of contaminated liquid.

Claims

1-50. (canceled)

51. A method of making an adsorbent material comprising mixing first particulate material with a second material, homogenising the mixture of the first and second materials, incorporating an impregnating or coating material capable of carbonisation, and carbonising the mixture.

52. A method according to claim 51, wherein the impregnating or coating material is curable.

53. A method according to claim 52, wherein the mixture is cured prior to or simultaneously with carbonisation.

54. A method according to claim 51, wherein the second material and the impregnating or coating material capable of carbonisation are the same.

55. A method according to claim 51, wherein the first and second materials are different.

56. A method according to claim 51, wherein one or both of the first and second materials is carbonaceous.

57. A method according to claim 56, wherein the carbonaceous material is selected from graphite intercalation compounds, compressed expanded graphite, natural flake graphite, activated carbon, graphite, carbon black, carbon nanotubes, graphene, glassy carbons, and amorphous carbon.

58. A method according to claim 51, wherein the first material is powdered graphite and the second material is powdered activated carbon.

59. A method according to claim 51, wherein at least one of the first and second materials is non-carbonaceous.

60. A method according to claim 51, wherein the first and second materials are mixed in ratios from around 1:99 to around 99:1, from around 10:90 to around 90:10, from around 20:80 to around 80:20, from around 30:70 to around 70:30, from around 40:60 to around 60:40, and around 50:50, all by weight.

61. A method according to claim 51, wherein the impregnating or coating material is one or a mixture of a thermosetting resin, a thermoplastic, or a monomer.

62. A method according to claim 61, wherein the impregnating or coating material is selected from phenolic resins, furan resins, oxidised polystyrene, polyvinyl alcohol, polyacrylonitrile, polyvinylidene chloride, cellulose, epoxy resins, polystyrene, sucrose, and polymethylmethacrylate, preferably wherein the impregnating or coating material is polyfurfuryl alcohol.

63. A method according to claim 51, wherein the impregnating or coating material is cured by thermal treatment, use of a chemical initiator, or a combination of the two.

64. A method according claim 51, wherein the material is activated, preferably following carbonisation, preferably wherein activation is effected by chemical and/or physical means.

65. A method according to claim 64, wherein the chemical means of activation is an acid, a salt, and/or a base.

66. A method according to claim 65, wherein the chemical means is one or a combination of phosphoric acid, zinc chloride, potassium hydroxide, and sodium hydroxide.

67. An adsorbent material which is made by, or is obtainable by, the method of claim 51, preferably wherein the first material is powdered graphite, the second material is powdered activated carbon or carbon black, and the impregnating or coating material is furfuryl alcohol.

68. The use of the adsorbent material of claim 51 in the treatment of a contaminated liquid.

69. A method of treating the surface of an adsorbent material comprising passing a current through the adsorbent material in the absence of adsorbed contaminants, preferably wherein the adsorbent material is made by, or obtainable by, the method of claim 51.

70. An adsorbent material which is made by, or obtainable by, the method of claim 69.

71. A method of removing contaminants from a quantity of contaminated liquid comprising passing an electric current through an adsorbent material prior to contacting it with the contaminated liquid, contacting the adsorbent with the contaminated liquid, allowing the adsorbent material to adsorb contaminants from the contaminated liquid, and regenerating the adsorbent by passing an electric current through the adsorbent.

72. A method of removing contaminants from a quantity of contaminated liquid comprising passing an electric current through an adsorbent material prior to contacting it with the contaminated liquid, contacting the adsorbent with the contaminated liquid, allowing the adsorbent material to adsorb contaminants from the contaminated liquid, and regenerating the adsorbent by passing an electric current through the adsorbent, wherein the adsorbent material is made by, or obtainable by, the method of claim 51.

73. A method of treating the surface of an adsorbent material according to claim 71, wherein the adsorptive capacity of the surface of the material is enhanced by electrochemical treatment.

74. A method according to claim 73, wherein the enhancement is achieved in-situ prior to adsorption in the absence of an adsorbate and/or when an adsorbate is adsorbed to the surface of the adsorbent material.

Description

[0081] The invention will now be described by way of example and with reference to the accompanying figures wherein:

[0082] FIG. 1 is a graph of the relationship between the voltage and bed depth at a series of different currents and used to calculate the conductivity of the beds;

[0083] FIGS. 2a and 2b are scanning electron microscope (SEM) images of Nyex and Nyex impregnated with furfuryl alcohol;

[0084] FIG. 3 is a graph of the concentration of AV-17 over time used to estimate the adsorption equilibrium of a composite material comprising compressed expanded graphite (GEC) and granulated activated carbon (GAC);

[0085] FIGS. 4a and 4b show the isotherm of adsorption of AV-17 onto non-activated CEG-GAC composite adsorbent and the isotherm of adsorption of AV-17 onto activated CEG-GAC composite adsorbent respectively;

[0086] FIG. 5 shows the regeneration efficiencies of the CEG-GAC composite materials for the adsorption of AV-17 over four adsorption-regeneration cycles;

[0087] FIG. 6 is graph of the adsorption of AV-17 from solution by a composite material comprising graphite and activated carbon;

[0088] FIGS. 7a and 7b show the dynamic light scattering measurement showing the particle size distribution of powdered graphite and the particle size distribution of powdered activated carbon respectively;

[0089] FIGS. 8a, 8b, and 8c show the kinetic results of the uptake of resorcinol by PGPAC6040 (a composite material comprising around 60% by weight powdered graphite (PG) and about 40% by weight powdered activated carbon (PAC)), PGPAC5050 (a composite material comprising PG and PAC is approximately equal proportions by weight), and PGPAC4060 (a composite material comprising about 40% by weight PG and about 60% by weight PAC);

[0090] FIGS. 9a, 9b, and 9b show the kinetic results of the uptake of AV-17 by PGPAC6040, PGPAC5050, and PGPAC4060;

[0091] FIG. 10 shows the adsorption efficiencies of the PGPAC variants over five cycles;

[0092] FIG. 11 shows the turbidity over time of samples of water containing samples of NYEX PGPAC4060, PGPAC5050, and PGPAC6040 in which the samples were stirred at 600 rpm;

[0093] FIG. 12 shows the turbidity over time of samples of water containing samples of NYEX PGPAC4060, PGPAC5050, and PGPAC6040 in which the samples were stirred at 800 rpm

[0094] FIG. 13 shows the turbidity over time of samples of water containing samples of NYEX PGPAC4060, PGPAC5050, and PGPAC6040 in which compressed air at 2 barg at a flow rate of 2 litres per minute was passed through the water; and

[0095] FIGS. 14a, 14b, 14c, and 14d show the turbidity of samples of water for four adsorption/regeneration cycles.

REFERENCE EXAMPLE 1

Prior Art MaterialNYEX

[0096] In order to provide a reference against which the materials produced by the method of the present invention could be compared, the physical characteristics of the adsorbent NYEX were determined. Details of the nature of NYEX may be found in GB2442950, the content of which is hereby incorporated by reference in its entirety. NYEX is produced by Arvia Technology Limited.

[0097] The surface area of the NYEX was measure by nitrogen adsorption to be 1.2 m.sup.2g.sup.1. As will be appreciated, this is considerably less than the surface area for most activated carbon adsorbents, which have surface areas of in excess of 1200 m.sup.2g.sup.1.

[0098] The NYEX was measured to have no internal porosity, with the bulk and particulate densities measured as 0.5 to 1.0 g cm.sup.3, and around 1.9 to 2.2 g cm.sup.3 respectively, depending on particle size, particle shape and size distribution. The free particle settling velocity and hindered particle settling velocity were estimated to be 312 cm min.sup.1 and 102 cm min.sup.1 respectively. These values were verified experimentally as 267 cm min.sup.1 and 89 cm min.sup.1, which were comparable to over 85% confidence with each other.

[0099] The electrical conductivity of a bed of NYEX was measured to be 0.240.03.sup.1 cm.sup.1. The voltage versus depth plots and the conductivity data used to calculate the bed conductivity are shown in FIG. 1 and Table 1 respectively.

TABLE-US-00001 TABLE 1 Current Potential drop Resistivity Conductivity (mA) (V cm.sup.1) ( cm) (.sup.1 cm.sup.1) 50 0.0282 3.4728 0.2879 100 0.0729 4.4888 0.2228 200 0.1461 4.4981 0.2223 400 0.2863 4.4072 0.2269 500 0.3537 4.3558 0.2296 800 0.5006 3.8531 0.2595 1000 0.7298 4.4938 0.2225

[0100] Granular activated carbon was measured to have a bed electrical conductivity of 0.080.01.sup.1 cm.sup.1. As a further comparison, the conductors copper and stainless steel have electrical conductivities of 610.sup.6.sup.1 cm.sup.1 and 1.510.sup.6.sup.1 cm.sup.1 respectively, whereas de-ionised water and sea water have electrical conductivities of 5.510.sup.8.sup.1 cm.sup.1 and 4.810.sup.2.sup.1 cm.sup.1 respectively. As can be seen from these relative values, NYEX is a relatively good conductor of electricity which means it may be used in electrochemical regeneration.

[0101] The adsorption kinetics of NYEX were investigated in respect of resorcinol and the dye acid violet-17 (AV17) by fluidising 50 g of NYEX in a litre of solution of resorcinol or AV17 and measuring the concentration of the respective adsorbent. It was found that adsorption equilibrium was reached at around 45 minutes. As such, a time of 45 minutes was used for the adsorption experiments with NYEX as this had been shown to be long enough to achieve adsorption equilibrium.

[0102] Kinetic modelling of NYEX has indicated that adsorption is best described by a pseudo-second order model which is based on a combination of physical and chemisorption between the adsorbates and the adsorbent. This is considered likely as NYEX has been measured to be non-porous by a combination of nitrogen adsorption and mercury porosimetry, and has also been shown to contain surface functionalities which affect the rate of adsorption onto the material. In addition, adsorption isotherms of NYEX show that the adsorption is monolayer adsorption on a non-porous adsorbent.

[0103] The regeneration efficiency of NYEX was also measured over four adsorption and electrochemical regeneration cycles. The results indicated that NYEX maintained high regeneration efficiencies over the cycles.

[0104] Adsorbents are usually exposed to extreme chemical, mechanical and thermal forces. Material disintegration is one of the main contributing factors to material losses in adsorption systems. For example, in steam regeneration processes for activated carbon adsorption systems, material loss of about 15% is directly linked to the elevated pressures used in the process. It is believed that NYEX undergoes material degradation and attrition due to the mechanical action of the air injected into the adsorption system which provides agitation and mixing. This creates fines which may reduce the quality of the treated water and limits the adsorption capacity. Turbidity measurements were taken and are discussed in more detail below. However, it was found that NYEX does undergo a degree of material attrition when in continuous use as an adsorbent.

REFERENCE EXAMPLE 2

Compressed Expanded Graphite

[0105] This reference example relates to a method of increasing the surface area of NYEX to improve the surface area and thereby increase the adsorption capacity.

[0106] In consideration of the relatively low surface area of NYEX, it was investigated whether the adsorption capacity could be increased by making some of the internal surface available. This was investigated by exfoliating the NYEX which involves heating the material at a high temperature for a short time, for example 800 C. for one minute. The volatisation of the intercalated species can result in a three hundred fold expansion in the volume of the material.

[0107] To synthesise a material of this nature, a sample of NYEX was sieved to homogenise the particle size. NYEX has a particle size of around 500 microns, but there is dust present with a size of less than or equal to 140 microns. As such, the sample of NYEX was size classified using a sieve tower to remove the dust. Particle sizes in the range of 140 to 425 microns were chosen as this represented the largest size range, which ensured homogenous exfoliation.

[0108] 5 g of the material were spread out into a stainless steel furnace tray and put into a muffle furnace preheated to 800 C. for one minute. The tray was left to cool in ambient conditions.

[0109] A comparison of the regular and exfoliated NYEX showed a significant particle size increase. The expanded graphite particles were compressed into compressed expanded graphite (CEG) particles with a similar morphology as the original unexpanded particles, but with less density due to the increased porosity. Dynamic Light Scattering (DLS) was used to measure the particle size of the expanded material, which found a mean particle size of 739 microns.

[0110] The surface area of the CEG particles was measured by nitrogen adsorption to be 22.5 m.sup.2 g.sup.1, which is a significant increase over the surface area of the NYEX material. The surface area of the expanded graphite before compression was measured to be 149 m.sup.2 g.sup.1. The bulk density of the CEG materials was measured as 0.19 g cm.sup.3 (compared to 0.53 g cm.sup.3 for NYEX) and the particulate density was measured using helium pycnometry as 0.12 g cm.sup.3 (compared to 1.91 g cm.sup.3) for NYEX. The CEG particles were measured by nitrogen adsorption to contain mesopores and macropores.

[0111] The low density of the material may make it unsuitable for use as an adsorbent in liquid systems, including the Arvia system for the removal of contaminants from a liquid as described in GB2495701, the contents of which is hereby incorporated by reference in its entirety. In particular, since the GEC particles float, there may be inefficient contact between the adsorbent and the adsorbates, and there is significant loss of adsorbent due to overflow of the treatment tanks. In addition, poor settling velocity (which may in fact be negative since the material floats on water) makes regeneration of the adsorbent difficult.

[0112] The average dry bed electrical conductivity of the CEG adsorbent was measured as 0.170.01.sup.1 cm.sup.1. This is less than the conductivity of the NYEX material, but still sufficient to allow electrochemical regeneration.

[0113] Since the CEG adsorbent floats on water, it proved difficult to conduct adsorption kinetics studies. However, adsorption equilibrium was believed to have been reached after around 60 minutes. Adsorption isotherm studies for the removal of AV-17 were carried out and the results show that the CEG material had a saturation adsorption capacity of about 7 mg g.sup.1, which is around double that of NYEX.

[0114] The regeneration efficiency of the CEG material was measured, but it was necessary to drain most of the liquid from the test rig due to the tendency of the CEG particles to float. The cell potentials for the CEG particles was measured to be more than twice the required cell potentials of NYEX to supply a current of 1 Amp required for regeneration. As such, the electrical energy cost for regeneration of the CEG particles is double that of NYEX The average regeneration capacity of the CEG particles was calculated to be 97% across four adsorption/regeneration cycles.

[0115] As such, although the CEG material offered improved surface area over NYEX and double the adsorptive capacity, the low density of the particles meant that they did not form a bed when used in an aqueous system. Therefore, CEG materials by themselves are not considered suitable for use in the Arvia process. However, in view of the large surface area and low density, they may find use in a packed column of material through which contaminated gases could be passed.

EXAMPLE 3

Compressed Carbonised Impregnated Nyex (CCIN)

[0116] This example relates to a method in accordance with the first and fourth aspects of the present invention, as well as materials according to the second and third aspects of the present invention.

[0117] In order to overcome the difficulties posed by CEG materials, in particular the low density, a composite material of the CEG material and furfuryl alcohol derived carbon was investigated.

[0118] This material was produced by using CEG as a substrate on which carbon was grown. The carbon growth was in the form of pyrolysis of polymerised furfuryl alcohol, which has been impregnated onto the CEG. Energy-dispersive x-ray spectroscopy indicated a significant change to the surface of the material as a result of the impregnation.

[0119] The NYEX particles were sieved as described in Example 2 and then mixed with furfuryl alcohol. The furfuryl alcohol was polymerised and the resulting material was simultaneously carbonised and exfoliated since the elevated temperature at which carbonisation occurred was enough to exfoliate the NYEX particles. This was followed by compression with a force of 15,000 kg and size reduction. The compressed material was then crushed to form particles, which were measure to have an average particle size of 720 microns. Scanning electron microscope (SEM) images of Nyex and Nyex impregnated with furfuryl alcohol are shown in FIGS. 2a and 2b respectively.

[0120] A variety of CCIN materials were produced were developed by varying the mass ratio of furfuryl alcohol (FA) to CEG; the mass ratio of activating agent (KOH) and CEG; and the compression force. Details of the various materials are shown in Table 2

TABLE-US-00002 TABLE 2 Particle Bulk Density Compression Density Material (g cm.sup.3) Force (kg) FA:CEG KOH:CEG (g cm.sup.3) Porosity CCIN 1 0.796 10,000 3:2 3:1 0.22 0.724 CCIN 2 1.094 10,000 2:1 3:1 0.27 0.753 CCIN 3 1.161 10,000 2:1 1:1 0.29 0.750 CCIN 4 1.200 15,000 2:1 1:1 0.39 0.675

[0121] The average surface area of the CCIN materials was measured by nitrogen adsorption as 24 m.sup.2 g.sup.1, which is greater than the surface area of NYEX, but only marginally greater than the surface of the previous CEG material. The surface area of the furfuryl alcohol-derived carbon was measured as 0.254 m.sup.2 g.sup.1, which suggests an insignificant contribution to the overall surface area of the material.

[0122] The densities of the various CCIN materials produced varied as shown in Table 2. The use of a higher proportion of furfuryl alcohol resulted in an increase composite density due to the formation of more furfuryl alcohol-derived carbon. Increasing the amount of KOH used during activation resulted in a lower density due to the increased porosity of the material. Further, extra compression forces increased the density due to increased compactness of the material.

[0123] The bed electrical conductivities of the CCIN composites were measured, and the results are shown in Table 3.

TABLE-US-00003 TABLE 3 Current Potential drop Resistivity Conductivity (mA) (V cm.sup.1) ( cm) (.sup.1 cm.sup.1) 50 0.053 6.53 0.15 100 0.106 6.53 0.15 200 0.209 6.43 0.16 400 0.417 6.42 0.16 800 0.826 6.36 0.16 1000 0.950 5.85 0.17

[0124] The average electrical conductivity was measured to be 0.160.01.sup.1 cm.sup.1, which is comparable to NYEX. This suggests that these CCIN composites would be able to undergo electrochemical regeneration.

[0125] The adsorption kinetics of the CCIN materials for removal of AV-17 dye was investigated. Whilst the CCIN particles have increased surface area available for adsorption, the kinetics were found to be slower. Without wishing to be bound by scientific theory, it is believed that treating NYEX to make some of the internal surface available for adsorption increased the porosity of the material. Consequently, adsorption was no longer limited to the external surfaces and the intra diffusion part of the adsorption process becomes the rate determining step. The amount of AV-17 removed from solution was greater than a comparative sample of NYEX.

[0126] Adsorption isotherms for the removal of AV-17 dye and resorcinol by the CCIN composites indicated that the CCIN materials adsorbed 6 mg g.sup.1 and 18 mg g.sup.1 for AV-17 and resorcinol respectively. This is around double the adsorption capacity for NYEX, which was measured at 3.5 mg g.sup.1 and 6 mg g.sup.1 for AV-17 and resorcinol respectively. Without wishing to be bound by scientific theory, it is believed that the larger size of the AV-17 dye molecules occupy a greater proportion of the surface of the adsorbent than the smaller resorcinol molecules.

[0127] A comparison of the regeneration efficiencies of the CCIN materials and NYEX is shown in Table 4.

TABLE-US-00004 TABLE 4 Charge Regeneration Efficiency (%) Density (C g.sup.1) Nyex CCIN 1 CCIN 2 CCIN 3 CCIN 4 8.6 100.2 63.5 64.2 65.8 65.7 12.9 101.2 76.1 75.8 76.4 76.8 17.1 100.6 87.3 88.1 88.6 88.6 21.4 101.4 100.1 100.2 100.7 100.4

[0128] Although the CCIN material showed comparable regeneration efficiencies as NYEX, they came at significantly increased energy costs as the low density nature of the CCIN particles resulted in less compact bed formation and consequently high electrical resistance across the bed.

[0129] At the same current densities as used with NYEX, the regeneration efficiencies of the CCIN materials were measured at around 63%, which is significantly lower than the efficiencies seen with NYEX, although the charge per unit pollutant removed was lower. Therefore, although the CCIN adsorbed between two and three times the amount of organics adsorbed by NYEX, the regeneration efficiency of 63% represents a lower charge per unit pollutant removed when compared to NYEX's 100% regeneration at 8.6 C/g of adsorbent.

EXAMPLE 4

CEG and Granular Activated Carbon (GAC) Composite

[0130] This example relates to a method in accordance with the first and fourth aspects of the present invention, as well as materials according to the second and third aspects of the present invention.

[0131] The composite of CEG and carbon (CCIN) resulted in a material of low density and an uneven growth of carbon on the CEG matrix. As such, a composite of CEG and similarly sized granular activated carbon (GAC) using furfuryl alcohol as an impregnating material to bind the CEG and GAC particles together was produced.

[0132] NYEX was exfoliated as described previously to obtain expanded graphite which was compressed into CEG. The CEG was size reduced by crushing and mixed with GAC. The mixture of CEG and GAC was impregnated with furfuryl alcohol followed by polymerisation, as described previously. In particular, a mixture of 50 g of CEG and 50 g of granular activated carbon was impregnated with 100 grams of furfuryl alcohol, and then polymerised with HCl and heat. The material was then carbonised and activated as described previously.

[0133] The surface area of the resulting material was measured by nitrogen adsorption and analysed by BET surface area model as 16 m.sup.2 g.sup.1. The density of the composite material was measured to be 1.39 kg m.sup.3 by helium pycnometry, and the particles were measured to have an average particle diameter of 766 microns. The free particle settling velocity and the hindered settling velocity were estimated to be 159 cm min.sup.1 and 48 cm min.sup.1 respectively.

[0134] The dry bed electrical conductivity of the CEG-GAC composite material was measured to be 0.080.01.sup.1 cm.sup.1. This is significantly lower than the conductivity of NYEX.

[0135] As shown in FIG. 3, kinetics experiments with AV-17 showed that this composite material reached adsorption equilibrium after 140 minutes, which is slower than the other materials tested and suggests greater porosity of the material.

[0136] The adsorption isotherm of the CEG-GAC composite material showed a maximum loading of 16 mg g.sup.1, which is around a four and a half times increase of the capacity of NYEX. This is a significant improvement in adsorption capacity and represents excellent performance as an adsorbent, despite the longer adsorption equilibrium time.

[0137] Two forms of the adsorbent were investigated; a non-activated one and an activated version. The performance of these two versions as adsorbents was tested using AV17 as the model pollutant. FIGS. 4a and 4b show the results of the adsorption experiments relating to these materials. In particular, FIG. 4a shows the isotherm of adsorption of AV-17 onto non-activated CEG-GAC composite adsorbent and FIG. 4b shows the isotherm of adsorption of AV-17 onto activated CEG-GAC composite adsorbent. The activated material adsorbed about three times as much as the non-activated material. These demonstrate the importance of activation in increasing the adsorption capacity of the adsorbent materials.

[0138] Regeneration efficiency experiments were carried out on the CEG-GAC composite materials. Four regeneration cycles were carried out and the regeneration efficiencies were significantly less than NYEX. FIG. 5 shows the regeneration efficiencies of the CEG-GAC composite materials for the adsorption of AV-17 over four adsorption-regeneration cycles. Lower bed conductivities generally lead to higher cost of regeneration. Lower regeneration efficiencies may be due to a lower charge per unit mass of pollutant adsorbed.

[0139] Materials having different ratios of GAC and CEG were produced, and the GAC was replaced with powdered activated carbon, but without significant improvements.

EXAMPLE 5

Graphite and Activated Carbon

[0140] This example relates to a method in accordance with the first and fourth aspects of the present invention, as well as materials according to the second and third aspects of the present invention.

[0141] Carbonised Natural Flake Graphite was produced by using Natural Large Flake Graphite (NLFG) as a substrate upon which activated carbon derived from polymerised furfuryl alcohol was grown. The method was similar to that used to produce CCIN, with the NLFG replacing the NYEX. In particular, the NLFG was sieved and then impregnated with furfuryl alcohol. The furfuryl alcohol was then polymerised and the material was subsequently carbonised. The carbonised material was activated using KOH and then crushed. The NLFG activated carbon composite was washed by rinsing with distilled water until the pH was between 6 and 7.

[0142] During kinetics experiments, this material was observed to undergo attrition and so the testing of this material was limited to the adsorption kinetics. FIG. 6 shows a graph of the adsorption of AV-17 from solution by this composite adsorbent. The results showed that this material reached adsorption equilibrium after 210 minutes, which is characteristic of a porous material.

[0143] Following on from the composite formed by forming activated carbon by pyrolysis of polyfurfuryl alcohol on a graphite flake substrate, an alternative composite material was produced. In the alternative material, activated carbon particles were bound to the graphite flakes.

[0144] This material was produced by mixing NLFG with GAC, impregnating the mixture with furfuryl alcohol, polymerising the furfuryl alcohol, carbonising the resulting mixture, activating the composite material with KOH, crushing the activated composite, and washing the activated composite. Upon crushing, the material was observed to disintegrate. Increased ratios of furfuryl alcohol were used in further composites to try to increase the binding strength, but no significant increase in composite strength was observed.

[0145] In order to increase the binding strength, the GAC was replaced with powdered activated carbon. However, when the resulting material was characterised and tested, it was found that the coverage of the flake graphite particles by the powdered activated carbon was insufficient and irregular, and contributed to the low stability of the material.

EXAMPLE 6

Powdered Graphite (PG) with Powdered Activated Carbon (PAC)

[0146] This example relates to a method in accordance with the first and fourth aspects of the present invention, as well as materials according to the second and third aspects of the present invention.

[0147] Previous attempts to produce a composite material of PG and PAC have not been successful. However, the methods of the first and fourth aspects of the present invention allow a composite of PG and PAC to be produced. The PG and PAC are mixed together and then impregnated/coated with furfuryl alcohol. The furfuryl alcohol is then polymerised and the mixture is subsequently carbonised. THE PG-PAC composite is then crushed and optionally washed.

[0148] The carbonisation is undertaken under inert conditions, for example in a stream of nitrogen gas. The mean particle sizes of the PG and PAC are similar to enhance bonding. The PG and PAC were analysed on a Malvern DLS Mastersizer to measure their mean particle size distribution, and the results of these tests are shown in FIGS. 7a and 7b. In particular, FIG. 7a shows the dynamic light scattering measurement showing the particle size distribution of powdered graphite and FIG. 7b shows the particle size distribution of powdered activated carbon. The PG was measured to have a mean particle size of 16 microns and the PAC was measured to have a mean particle size of 18 microns. The surface areas of the powdered graphite and the powdered activated carbon were measure by BET nitrogen adsorption as 0.7 m.sup.2 g.sup.1 and 1270 m.sup.2 g.sup.1 respectively.

[0149] Three versions of the PG-PAC composite were developed. In the first version, an equal mass of PG and PAC were combined and will be referred to as PGPAC5050. Two other versions were also produced in which the PG and PAC were combined in ratios of 40:60 by mass and 60:40 by mass and will be referred to as PGPAC4060 and PGPAC6040.

[0150] The PGPAC4060 was developed with a view to increasing absorption capacity, and PGPAC6040 was developed with a view to increasing the regeneration ability.

[0151] Table 5 provides a summary of the surface areas, porosities, and settling velocities of the three PGPAC adsorbents.

TABLE-US-00005 TABLE 5 PGPAC4060 PGPAC5050 PGPAC6040 BET Surface Area (m.sup.2 g.sup.1) 116.8 5.1 24.3 0.9 19.1 1.3 Micropore surface area 101.9 18.2 15.9 (m.sup.2 g.sup.1) External surface area (m.sup.2 14.8 6.3 3.2 g.sup.1) Micropore contribution to 87.2 74.9 83.2 total surface area (%) Adsorbent bed porosity 63.5 76.1 87.3 Density (kg m.sup.3) 1550 1680 1805 Particle settlement 205.0 238.9 269.1 velocity (cm min.sup.1) Hindered settlement 63.6 75.5 86.2 velocity (cm min.sup.1)

[0152] The effect of the ratio of the constituents had a large effect on the surface area, and therefore adsorptive capacities, of the composite materials. In particular, the composite with the greater ratio of PG to PAC had greater densities and settling velocities, although the differences in these properties were less significant than the difference in surface areas. As such, the gain of surface area shows a greater proportional increase than the gains in density and settling velocity.

[0153] These materials were measured to contain micropores and mesopores. The contributions of micropore surface areas and external surface areas are shown in Table 5, which shows that the PGPAC composite materials have high proportions of their surfaces as being micropores. These materials may therefore be classified as mainly microporous with between 17 and 25% of the surface area contributed by mesopores.

[0154] The electrical conductivity of a bed of each of the PGPAC variants was measured. The results show a positive correlation between the amount of powdered graphite in the composite and the electrical conductivity. PGPAC4060, which has the lowest amount of powdered graphite, was measured to have a bed conductivity of 0.280.02.sup.1 cm.sup.1; PGPAC5050 was measured to have a bed conductivity of 0.320.01.sup.1 cm.sup.1 and PGPAC6040 was measured to have a bed conductivity of 1.850.05.sup.1 cm.sup.1.

[0155] This example demonstrates how the methods of the present invention can be used to produce tailored adsorbents. For example, where an adsorbent with a high adsorption capacity is required, the method can be used to produce an adsorbent with a high surface area, such as PGPAC4060, whereas if an adsorbent with good electrical conductivity is required, the method can be used to produce an adsorbent with high conductivity, such as PGPAC6040.

[0156] The adsorption kinetics of the three PGPAC variants were investigated. 50 g of the PGPAC variants was mixed into one litre solutions of AV-17 and resorcinol at concentrations of 100 ppm and 1000 ppm respectively, and the uptake was measured regularly. The kinetic results of the uptake of resorcinol by PGPAC6040, PGPAC5050, and PGPAC4060 are shown in FIGS. 8a, 8b, and 8c respectively. It was found that adsorption onto these PGPAC composites was slower than NYEX, which reached equilibrium after around 45 minutes. The PGPAC composite materials had not reached equilibrium after 300 minutes of adsorption. It is believed that this is a result of the larger surface area available for adsorption as well as the increased porosity of the PGPAC composite materials.

[0157] As shown in FIGS. 9a, 9b, and 9c, the uptake of AV-17 was similar to that of resorcinol by PGPAC6040, PGPAC5050, and PGPAC4060 respectively.

[0158] Although the PGPAC materials have longer adsorption equilibrium times, the rate of uptake of the resorcinol and AV-17 was significantly higher than the rate for NYEX. FIGS. 8a to c and 9a to c show that the PGPAC composites have significantly greater uptake rates than NYEX. In particular, after 60 minutes each of the PGPAC variants had adsorbed considerably more adsorbate than NYEX. As such, it does not appear necessary to wait until adsorption equilibrium has been reached to regenerate the adsorbent material.

[0159] The adsorption capacities for AV-17 were measured as 3.5 mg g.sup.1, 7.3 mg g.sup.1, 9.7 mg g.sup.1, and 13.8 mg g.sup.1 for NYEX, PGPAC6040, PGPAC5050, and PGPAC4060 respectively. The adsorption capacities for resorcinol were measured as 8.6 mg g.sup.1 , 52.8 mg g.sup.1, 78.3 mg g.sup.1, and 100.1 mg g.sup.1 for NYEX, PGPAC6040, PGPAC5050, and PGPAC4060 respectively.

[0160] NYEX has been measured to have a sustained regeneration efficiency of over 100% over 5 cycles. The adsorption capacity of the NYEX material appears to improve after use, which is reflected in the efficiency in excess of 100%. Without wishing to be bound by scientific theory, it is believed that the passage of current through the adsorbent alters the surface properties of the adsorbent which increases its absorptive capacity. FIG. 10 shows the adsorption efficiencies of the PGPAC variants over five cycles. This shows that surface modification of the NYEX can be achieved in situ as during treatment the contaminant surface coverage reduces, resulting in the modification of the surface enhancing adsorption. The averages regeneration efficiencies of the PGPAC were calculated to be 97.5%, 95%, and 100% for PGPAC5050, PGPAC4060, and PGPAC6040 respectively. The left hand bar in each group of three bars relates to PGPAC5050, the centre bar relates to PGPAC4060, and the right hand bar relates to PGPAC6040.

[0161] The stability of the PGPAC composite material variants was investigated since it has been observed that NYEX may undergo attrition over prolonged use.

[0162] FIG. 11 shows the results of bench scale isotherm experiments carried out in 500 ml conical flasks. A 4 g sample of NYEX was mixed with 100 ml of water and stirred continuously at 600 rpm using an ER Lauda magnetic stirrer. Samples of the water were taken every 30 minutes to be tested for turbidity. The samples were left to settle for ten minutes after being taken to allow the larger particles to settle, leaving the colloidal supernatant to be analysed for turbidity. Ten samples were taken at each measuring to allow for the calculation of a mean measurement.

[0163] The results indicate a steady increase in turbidity for the NYEX sample until 120 minutes after which only a marginal increase was noted. It is clear from FIG. 11 that each of the PGPAC composites underwent significantly less attrition compared with NYEX particles.

[0164] Further experiments to investigate the effect of the stirring rate on the attrition of the NYEX and PGPAC particles. These experiments revealed that the rate of attrition increased with stirring rate up to around 800 rpm after which it decreased slightly. Again, the rate of attrition of the NYEX particles was higher than that of the PGPAC particles. The results of these experiments are shown in FIG. 12.

[0165] Still further experiments were carried out to investigate the effect of passing compressed air through the system. In one arrangement of the Arvia process, compressed air is passed through the bed of adsorbent and through the liquid to provide mixing and oxygen. In the present experiments, 100 g of adsorbent was mixed into 1 litre of water using compressed air at 2 barg at a flow rate of 2 litres per minute and samples were taken every 30 minutes for 210 minutes. As before, the samples were allowed to settle for ten minutes before the supernatant was collected and the turbidity measured. FIG. 13 shows the results of these experiments. There was a significant increase in fines during the initial stage of mixing followed by saturation in the rate of formation of the fines. Again, the PGPAC materials showed improved resistance to attrition when compared with NYEX.

[0166] The effect of current passed through the adsorbent materials during electrochemical regeneration on the stabilities of the adsorbents was investigated by carrying out mock adsorption and regeneration without any adsorbates in the water.

[0167] A test cell was filled with 100 g of adsorbent material and 1 litre of water. For the first mock adsorption, air was mixed into the system to replicate the mixing of the particles in water for 30 minutes. After 30 minutes, the air was stopped and the water was left to stand for 10 minutes to allow the solid adsorbent particles to settle in the regeneration zone. A mock regeneration was then carried out by passing a current of 1 A through the bed of adsorbent particles for 20 minutes at a charge density of 12 C g.sup.1. The active electrode area was 70 cm.sup.2 and the cathode used was 0.3% acidified NaCl solution. The mock adsorption and regeneration were repeated over four cycles and the results are shown in FIGS. 14a, b, c, and d.

[0168] The results from the mock adsorption and regeneration experiments show a general trend which suggests that electrochemical treatment contributes to the attrition of the adsorbent. There was also a gradual increase in turbidity, which suggests that electrochemical regeneration increases attrition of the adsorbent.

[0169] In summary, the methods of the present invention allows for the production of adsorbent materials, the properties of which can be attuned to the particular environment in which the adsorbent materials are to be used. In particular, the methods may be used to produce an adsorbent material with a high adsorption capacity, or to produce an adsorbent material which has high conductivity. The adsorbent materials produced by the methods of the present invention may be treated in accordance with the fifth aspect of the present invention.