Activated Carbon Ullman

Carbon, 5. Activated Carbon K LAUS LAUS-DIRK  HENNING , CarboTech Aktivkohlen GmbH, Essen, Germany HARTMUT VON K IENLE IENLE, (formerly Degussa AG, Hanau), Heusenstamm, Germany

1. 2. 2.1. 2.1. 2.2. 2.2. 2.3. 2.3. 2.4. 2.4. 2.5. 3. 3.1. 3.2. 3.3. 3.3. 3.4. 3.4. 3.4. 3.4.1. 1. 3.4. 3.4.2. 2. 3.5. 3.5. 3.6. 3.6. 3.7. 3.7.

General Aspects. . . . . . . . . . . . . . . . . . Carb arbonac onaceeous ous Ads Adsorb orbents ents . . . . . . . . . . Types ypes of Carb Carbon onac aceo eous us Adso Adsorb rben ents ts . Chemi emical cal Prop Proper erti tiees. . . . . . . . . . . . . Mecha echan nical cal Prope opertie rtiess . . . . . . . . . . . Adsor sorption tion Prop Proper erti tiees . . . . . . . . . . . Quality Control . . . . . . . . . . . . . . . . Production . . . . . . . . . . . . . . . . . . . . . . General Aspects . . . . . . . . . . . . . . . . Raw Materials . . . . . . . . . . . . . . . . . Activ ctivat atin ing g Fur Furnac naces. . . . . . . . . . . . . Meth ethods ods of of Ac Activ tivati ation . . . . . . . . . . . Chem Chemic ical al Acti Activ vatio ation n . . .. . . .. . . .. . Gas Gas Acti Activ vatio ation. n. . . . . . . . . . . . . . . . . . Gran Gr anul ular ar and and Pelle elleti tize zed d Carb Carbon onss . . Carbo arbon n Molec olecu ular lar Siev ieves . . . . . . . . . Furth urther er Treat eatment. ent. . . . . . . . . . . . . .

1 2 2 5 5 6 7 9 9 10 11 13 13 13 14 15 15

1. Gen Genera erall Aspects Aspects  Definition. Activ Activate ated d car carbon bon [7440[7440-4444-0] 0] is the collec collectiv tivee name name for car carbon bonace aceous ous adsor adsor-bents which are defined as follows [1]: Activated carbons are nonhazardous, processed carbonace naceou ouss ma mate teri rials als havi having ng a poro porous us stru struct ctur uree and and a large internal surface area. They can adsorb a wide variety of substances, that is, they are able to attract molecules to their internal surface and therefore act as adsorbents. The pore volume of  acti activa vate ted d carb carbon onss is gene genera rally lly grea greate terr than than 0.2 mL/g. The internal surface area is generally greater than 400 m2 /g. The width of the pores ranges ranges from from 0.3 to sever several al thousa thousand nd nanome nanometers ters.. All activ activate ated d car carbon bonss [2–6] [2–6] are charac character terized ized by a ramified pore system (Fig. 1) in which pores of various sizes, such as mesopores (d  ¼ 2–50 2–50 nm), nm), micr microp opor ores es (d  ¼ 0.8–2.0 nm) and submic submicrop ropore oress (d   0.8 0.8 nm) nm) bran branch ch off  off  from macropores (d  ! 50 nm). X-ray investigations show that the carbon is mainly in the form of very small crystallites with a graphite structure. However, the typical

Ó 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 10.1002/14356007.n05_n04

3.8. 3.8. 4. 4.1. 4.1. 4.1. 4.1.1. 1. 4.1. 4.1.2. 2. 4.1. 4.1.3. 3. 4.1. 4.1.4. 4. 4.1. 4.1.5. 5. 4.2. 4.2. 4.2. 4.2.1. 1. 4.2.2. 4.3. 4.3. 4.4. 4.4. 5. 6.

Impregnat gnatio ion n. .. .. .. . .. .. .. .. .. Applications. . . . . . . . . . . . . . . . . . . . GasGas-Ph Phas asee Appli pplica cati tion onss . . . . . . . . . . Solv Solven entt Rec Recov over ery y . .. .. . .. .. .. . .. Proc Proces ess-G s-Gas as and and Air Air Puri Purifi fica cati tion on . . . . Gas Gas Sep Separ arat atio ion n . .. .. .. . .. .. .. . .. Gaso Gasoli line ne Vapor apor Adso Adsorp rpti tion on . . . . . . . . Flue Flue Gas Gas Cle Clean anin ing g. . . . . . . . . . . . . . . Liqu Liquid id-P -Pha hase se Appli pplica cati tion onss . . . . . . . . Water ater Treat reatme ment nt . . . . . . . . . . . . . . . . Micellaneous Liquid-Phase Appl Appliicati cation onss . . . . . . . . . . . . . . . . . . . Impr Impreg egna nate ted d Acti Activa vate ted d Carb Carbon on . . . . Cata Cataly lyst stss and and Cata Cataly lyst st Supp Suppor orts ts . . . Rege Regene nera rati tion on and and Reac Reacti tiva vati tion on . . . . . Economic Aspects . . . . . . . . . . . . . . .

16 16 16 17 18 19 20 20 21 21 22 23 24 24 25

graphi graphite te arrang arrangeme ement nt of the car carbon bon platel platelets ets one above above the other other is absent absent.. The crysta crystallin llinee region regionss are only 0.7–1.1 nm thick and 2.0–2.5 nm in diamet diameter er,, a consid considera erably bly smalle smallerr size size than than is observed in graphite (see Table 1). This means that in each crystallite there are usua usuall lly y only only thre threee or four four layer layerss of carbo carbon n atom atomss with with abou aboutt 20–3 20–30 0 carbo carbon n hexa hexago gons ns in each each layer. The spaces between the crystallites are filled with amorphous carbon which is bonded in three three dimensi dimensions ons with with other other atoms, atoms, especi especiall ally y oxygen. The irregular arrangement of the carbon atoms atoms is broken broken up by numerous numerous cracks cracks and fissures (pores), which are often idealized as cylindrical in shape. The large number of very fine fine pores pores (micro (micropor pores es and submicr submicropo opores res)) gives gives activ activate ated d car carbon bon a large large inner inner surfac surface, e, which is the basis of its remarkable adsorption properties. A common feature of these adsorbents is an amorphous amorphous structure structure with a high carbon content content and a hydrop hydrophob hobic ic surfac surfacee proper propertie ties. s. Activ Activate ated d carb carbon on is ma made de on comm commer erci cial al scal scalee from from carbon-con carbon-containin taining g raw materials materials (wood, (wood, peat,

2

Carbon, 5. Activated Carbon

Figure 1. Schematic model of activated carbon

coconut shells, lignite, hard coal) by chemical activation or gas activation. Activated carbon is commercially available in shaped (cylindrical pellets), granular, or powdered form.  History. The medical use of charcoal was described as early as 1550 B.C. in an ancient Egyptian papyrus and later by H IPPOCRATES and PLINY the ELDER. The products described at that time and later in the 1700s were of varying effec effectiv tivene eness ss and includ included ed wood, wood, blood, blood, and animal charcoals. The decolo decoloriza rizatio tion n of sugar sugar solutio solutions ns by bone black was first carried out commercially in England in 1811. Bone black, however, consists mainly of calcium phosphate and contains only a small amount of carbon. Strictly speaking, it is not a carbon. Methods of obtaining decolo decolorizi rizing ng charco charcoals als from from plant plant materi materials als were first set out in British patents in 1856– 1863. The first industrially manufactured activated carbons in the proper sense were Eponit decolo decolorizi rizing ng car carbon bons, s, which which have have been been proproduce duced d sinc sincee 1909 1909 acco accord rdin ing g to a pate patent nt of  R. VON OSTREJKO [7] by heating wood charcoal with with steam steam and and carb carbon on diox dioxid idee in a furn furnac acee specifica specifically lly designed designed for the purpose. purpose. In 1911 in the Netherlands, Norit NV began commercial activation of peat by using steam. The chemical activation of sawdust with zinc Table 1. Typical dimensions of crystal regions

Natural graphite Activated carbon

 L a, nm

L b, nm

c /2, nm

210 2.0–2.5

94 0.7–1.1

0.351 0.35–0.37

chloride to produce Carboraffin was first described in a patent of the Austrian Association for Chemical and Metallurgical Production [8]. This This proc proces esss was was firs firstt oper operate ated d in Auss Aussig ig,, Cz Czec ech h Republic, and in 1915 by Bayer. In the United States States during during World orld War I, the activ activati ation on of  coconut charcoal for gas masks was developed. The many and varied applications of activated carbon are amply illustrated by the existence of  over over 1500 1500 manufa manufactu cturin ring g patent patentss worldw worldwide ide [3]. [3]. Duri During ng the the firs firstt deca decade dess of this this cent centur ury y, activated carbon was used mainly for the purification of products of the chemical, pharmaceutica ceutical, l, and food food industr industries ies;; purif purificat ication ion of  drinking water was also an important application from the outset. It is increasingly used for the prevention of environmental pollution and for meeting meeting the constantly constantly increasing demands demands for purity of natural and synthetic products.

2. Carbonaceo Carbonaceous us Adsorbents Adsorbents 2.1. Types of Carbonaceous Carbonaceous Adsorbents Due to the wide variation in the properties of  activ activate ated d car carbon bon there there is no compre comprehen hensiv sivee nomenclature or standardization. The product group can be characterized by appearance, pore radius radius distributi distribution, on, or by typical typical applications applications.. Classification by appearance: * * *

Powdered activated carbon (PAC) Granular activated carbon (GAC) Cylindrical pellets

Carbon, 5. Activated Carbon * * *

Spherical pellets Activated carbon fibers (ACF) Activated coke Classification by pore radius distribution:

* *

*

Activated coke Activated carbon fine-pore medium-pore wide-pore Carbon molecular sieves (CMS) Classification by field of application:

* * * * * *

Carbon molecular sieve (CMS) Decolorizing carbon Water-treatment carbon Catalyst carbon Drinking-water carbon Solvent-recovery carbon

Carbonaceous adsorbents are usually characterized by appearance as powder, granules, or formed shapes such as cylindrical or spherical pellets. Grouping only according to end use is not meaningful because the requirements are so varied. Also no useful classification is possible on the basis of raw materials or production methods. In practice these products are often called, for example, powdered decolorizing carbon, granular carbon for water treatment, or pelletized catalyst carbon. In industrial practice, carbonaceous adsorbents are often classified by their pore radius distributions into activated carbon, activated coke, and carbon molecular sieves [9]. Some information on characteristic data and applications of commercial activated cokes and carbon molecular sieves are given in Tables 2, 3, and 4. Figure 2 shows the schematic pore diameter distribution of activated carbon, activated coke, and carbon molecular sieves. Carbon molecular sieve clearly exhibits narrower pores than activated carbon and activated coke. The pore diameters are matched to the molecular sizes of the gases to be adsorbed.   Activated carbon typically exhibits pore volumes of well above 25 cm3 /100 g and pore diameters of less than 2 nm. The specific inner

   s     t    n    e      b    r    o    s      d    a      d    e    s    a      b      n    o      b    r    a    c      f    o    a     t    a      d      l    a    c      i    p    y     t      d    n    a    s    n    o      i     t    a    c      i      l    p    p    a    e    s    a      h    p      s    a      G   .       2     e       l       b     a T

   s    e    v    e    m      i    s    o    y    r    r      f    r    a    e    v        4      l    u    o    c    c      H    e    e      l    r      C  ,    o    r        2      i    m      O    a    s    n      d    m   a    o    n    o    g    o      b    a      f    r      i    r      b        2    a      C      N

     0      0      0      1     0      0  .      0      2    a      1      6    c      2

     0      3  .      0      0      1     0      2  .      5      0      >     <     8

 ,    n    o      i     t      l    a    p    r    v    o    s    o      d    m    a    e    r    n     x    e    a      O      k    r    u     N    o      f    c      d     d      d    n    n    e    a    a     t    a    n        2    v      i      O      i    x     S     t    o    c      i      A      d

     0      0      6   –      0      0     0     0      0     0     9      5     9     1

     1  .      0     3  .   –     0     0      5   –     0      4     0      0  .      2  .      5      0     0      <     8

   e    r    o    p      e    y    r    e      d      i    v    w    o    c  ,    g    n    e    n    s    o    r      i      l    n      d     i      b    r    o    n    o     b    a    a      b    c   -    r    a    n      d    o     h    c    g    o    e      i     t      i     t    r    a    p      h     d    v    r      f    y      i    o     t    s    o     h    c      d      A    a

     0      0      0      2      0      2      4   –   –      0      0     0     0      0     0     1      3     6     2

     0      0      5      5  .      1  .      1   –      0     1     0   –   –     0     0      3  .      5  .      0     5      0     0     1     8

   e    r    o    p      m    u      i      d    e    m  ,    n    o      b    r    a    c      d    e     t    a    v      i     t    c      A

   n    o      i     t    p    r    o    s    g      d     i    n    a      l  ,      i    o    y    r      b    s    e   -    n    o    v    m     b    o    u    r    c      i    a    e    r      d    c     t    e    o    r    n    m     d    e      f    y    v      l    o     h    o    s

     0      0      0      2      5      2      4   –   –      0      0     0     0      5     0     1      3     7     2

     0      0      4      6  .      7  .      1   –      0     0     0   –   –     0     0      4      5  .  .      2     5      0     0     1     8

   e    r    o    p      e    n      i      f  ,    n    o      b    r    a    c      d    e     t    a    v      i     t    c      A

 ,    g      l    n    o      i    r      l     t      i    n    o     t    s    o    c      b      u    r    w    a    o    o      h     d     l    x    o     f    s    e  ,    o    n      d    p    n    o    n    u    o     b    a    n     i    r      0    a      0    r    a     t    c      0      i    p    e      2    r    r    o     0    a      l      2    c    o    e    r    s      d     5   –      k     i      d    y   –      0    a    a    a      h     0     0     0     t      0     0     1    n      i      4     8     2

     K    g    g     k      /    e    m        3      /        2      /    z      J      i    g     /    m    m  ,    s      k    g  ,     t    y    e    a      i     a  ,      k       3    r    e    o    y    r    c     t     a ,    m   p    a      i    g     /    g    a    p    y     /    s     t    r      /    g    o    e    a    n     i    c      f      L     L    c    e    s      k    a     t  ,    e    m   m     f      d    n    y    e     t    a    r      d     d     i    m   m   m   u    e    e     t    s    u    n    n    s      h     t    n     l    c    n    c    c    e    o    a    e      0     i      f      i      f      d    v     0    p    r      i    a    e    e      2     2     i    c    m   p    u    r      <     >    c    e    e    o    p    r    o    p    p      C     A     T     P       d        d      S     S

       3

   s    n    o      i     t    a    c      i      l    p     t    n    p    a    e      l      b    a    r    c    o    s      i      d    p    y      A      T

     0      0      2      7  .      5  .      1   –      0     0     0   –   –     0     0      5      3  .  .      0     5      0     0     1     8

3

 .    s    n    o      i     t    a     t      i    m      i      l      h     t      i    w    e      l      b    a    c      i      l    p    p    a    y      l    n    o  ,    n    o      b    r    a    c      d    e     t    a    v      i     t    c    a      d    e    r    e      d    w    o    p    r    o      F     a

4

Carbon, 5. Activated Carbon      l    a    v    o    m    e    r    e    e  ,    s      k    n    e    o      i    o    n    c     t    a    g    n      d    a    o    s    n    r      0    e      i    a      i     t    n    a    o    m     d     0    v    z      f    n     6   –      0      i     t    o    a      0     0     0    c    o    e      0     0     9      A      d      5     9     1

Table 4. Applications and characteristic data of activated carbon in catalysis

Adsorbent      1  .      0     3  .   –     0     0      5   –     0      4     0      0  .      2  .      5      0     0      <     8

Applications

   e    r    o    p      e      d      i    w  ,    n    o      b    r    a    c      d    e     t    a    v      i     t    c      A

   r    e     t    a    w    e     t    s    a    w  ,    n    n    o    o      i     t      i    a     t    a    z      i    c      i    r      f      i    o      l    r    o    u    c    p    e      d

     0      0      0      2      0      2      4   –   –      0      0     0     0      0     0     1      3     6     2

Compacted density, a kg/m3 Apparent density,a kg/m3 True density, kg/m3 Pore volume for pore size d  < 20 nm mL/g Pore volume for pore size d  > 20 nm mL/g Specific surface area, m2 /g Specific heat capacity, J/kgK 

     0      0      5  .      1  .      4      0     1     1   –   –   –     0      0      3  .      5  .      0     5      0     0     8     8

Activated carbon fine-pore

medium- and wide-pore

catalyst for chemical reactions 400–500

catalyst support for impregnation

800

600–700

2100–2200 0.5–0.7

2100–2200 0.3–0.6

0.3–0.5

0.5–1.1

1000–1200

1000–1500

850

850

300–450

a

For powdered activated carbon, only applicable with limitations.

   s     t    n    e      b    r    o    s      d    a      d    e    s    a      b      n    o      b    r    a    c      f    o    a     t    a      d      l    a    c      i    p    y     t      d    n    a    s    n    o      i     t    a    c      i      l    p    p    a    e    s    a      h    p        d      i    u    q      i      L

  .       3     e       l       b     a       T

   e    r    o    p      m    u      i      d    e    m  ,    n    o      b    r    a    c      d    e     t    a    v      i     t    c      A

   r    e     t    a    w    e     t    s    a    w    n    o      d     i    n     t    a    a      i    c      f    e      i      l      b    r    a    u     t    p    o    p

   e    r    o    p      e    n      i      f  ,    n    o      b    r    a    c      d    e     t    a    v      i     t    c      A

     d      l    n    o      i      l    o    a    g     t  ,    a    v    s    n    o     t      i    e      f    m   n    a      f    e    a    r     t    u  ,      l    c    e    n     l    o      d    o      i  ,     t    p      0    a    o    r    y    r      0    n    e    c      i      0      2      i    v    r      0      2    o    m   o      5   –      l    c   –      0      h     f    e    c    o    r      0     0     0    e      0     0     1      d      4     8     2

     0      0      0      2      5      2      4   –   –      0      0     0     0      5     0     1      3     7     2

   s    n    o      i     t    a    c      i      l    p     t    n    p    a    e      l      b    a    r    c    o    s      i      d    p    y      A      T

     0      0      6  .      7  .      3      0     0     1   –   –   –     0      0      4  .      5  .      0     5      0     0     8     8

     0      0      7      5  .  .      2      0     0     1   –   –   –     0      0      5  .      3  .      0     5      0     0     8     8

     K    g    g     k        3      /    e    m      /        2      /    m      i    z      J    g     /    m  ,    s      k    g  ,     t    y      k    e    a     a      i  ,      *       3    r    e    o    y    r    c     t  ,    m   p    a      i    g     /    g    a    p    y     /    s     t    r      /    g    o    e    a    n     i      L     L    c    c    e    s      k    a     t  ,      f      d    n    y    e    m   m     f    e     t    a    e      d     d     i    m   m   m   r    e     t    s    u    n    n    u    s      h     t    n     l    c    n    c      i    c    e    o     0     0     i    a    e      d    p    r      i      f      i      2     2     f    a    e    v    c    e    m   p    u    r      <     >    e    c    e    o    p    r    o    p    p      C     A     T     P       d        d      S     S

       3

surface area of commercial activated carbons ranges between 500 m2 /g and 1500 m2 /g. Due to its hydrophobic character, activated carbon is particularly suited to the adsorption of nonpolar organic substances, which has led to a broad range of applications in air pollution control and water treatment. In addition, catalytic reactions occur on activated carbon surfaces. For this reason activated carbon is used commercially as catalyst or catalyst support.  .    s    n    o      i     t    a     t      i    m      i      l      h     t      i    w    e      l      b    a    c      i      l    p    p    a    y      l    n    o  ,    n    o      b    r    a    c      d    e     t    a    v      i     t    c    a      d    e    r    e      d    w    o    p    r    o      F     a

 Activated coke is manufactured from lignite or hard coal and has typical pore volumes of up to 25 cm3  /100 g and specific surface areas of up to 400 m2 /g. Depending on the source material

Figure 2. Schematic pore diameter distributions of carbonaceous adsorbents

Carbon, 5. Activated Carbon

and the manufacturing process, adsorptive or catalytic characteristics may predominate. Activated coke is used for removing SO2 and dioxins from waste and flue gases. Carbon molecular sieves have micropore diameters that range from 0.50 to 1.00 nm with pore volumes of up to 20 cm3 /100 g. The separation effect of carbon molecular sieves (CMS) is based on differing rates of diffusion into the pore system. For this purpose the pore diameters of the CMS are matched carefully to the diameters of the molecules to be separated. Thus, a kinetic separation effect (sieve effect) is obtained. Larger molecules are adsorbed at a substantially lower rate in spite of higher equilibrium loads. For example, oxygen is adsorbed ten times faster than nitrogen although its molecular diameter is only slightly smaller.

2.2. Chemical Properties Activated carbon contains not only carbon, but also small amounts of oxygen, nitrogen, sulfur and hydrogen, which are chemically bonded in the form of various functional groups, such as carbonyl, carboxyl, phenol, lactone, quinone, and ether groups [6, 10, 11]. These surface oxides are sometimes derived from the raw material or they can be formed during or after the activation process by the action of air or water vapor. They usually have acidic character (seldom basic), and they give to the activated carbon the character of a solid acid or base. Under suitable conditions, surface sulfides and carbon–chlorine compounds can be formed. These surface chemical properties play a significant role in adsorption and catalysis. All the raw materials used for the production of activated carbons contain mineral components which become concentrated during the activation process. Moreover, the inorganic chemicals used in chemical activation are often only incompletely removed. The ash content of  many products is reduced by water or acid washing. Hence, commercial products contain from a few tenths of a percent up to 20% ash. The main constituents of the ash are salts of the alkali and alkaline earth metals, mostly carbonates and phosphates, together with silica, iron, and aluminum oxides.

5

2.3. Mechanical Properties Performance characteristics for activated carbon are generally expressed in relation to its mass. If figures on a volume basis are needed, bulk density or tapped density figures are also needed to carry out the conversion calculation. The bulk density is very much dependent on the filling technique, the geometry of the vessel used, and the grain size of the material. Therefore, a defined tapping or shaking process is usually included in the case of the higher tapped densities [1]. Another important feature of activated carbon is its grain size distribution. This determines the resistance of a layer of activated carbon to the flow of a liquid or gas (Fig. 3). The fineness of grinding of a powdered carbon affects the filtration properties. For filling a high adsorption tower or for any other static loading, the granules or pellets must be resistant to crushing. For transportation procedures such as pneumatic delivery, the abrasion resistance is very important. Care must also be taken that the thermal and chemical resistance properties are adequate to withstand any severe temperature variations or aggressive environments such as oxidizing atmospheres.

Figure 3. Pressure drop of molded activated carbon with different particle diameter (20  C, 1-m layer,dense packing)

6

Carbon, 5. Activated Carbon

Figure 4. Pore size distribution of different activatedcarbons —— Gas-phase activated carbon (narrow pores); . . . . . Liquid-phase activated carbon (wide pores)

2.4. Adsorption Properties The adsorption properties of activated carbon [2–6, 12–16] depend principally on its inner surface area, which in commercial products is 500–1500 m2 /g. To make use of the inner surface which is provided by the walls of the pores, the accessibility of this surface is important, that is, the pore size and the pore size distribution. This is determined by various methods, mostly from nitrogen adsorption isotherms, and is represented as integral or differential distribution curves (Fig. 4). The term “adsorption“ refers to the accumulation of gaseous or dissolved components on the surface layer of a solid (the adsorbent) [9, 11–15]. On their surface activated carbons have “active sites“ where the binding forces between the neighboring carbon atoms are not fully saturated and adsorption of foreign molecules takes place. Depending on the intensity of interactions between adsorbent and the component to be adsorbed, an adsorption enthalpy of  5–65 kJ/mol is released. With increasing temperature the quantity of the matter adsorbed in the equilibrium state decreases; in gas phase the influence of temperature on adsorption capacity is greater than in liquid phase. The reverse of the adsorption process is called desorption. The adsorption mechanisms are classified as chemisorption and physisorption. Physisorption is reversible and involves only physical interaction forces (van der Waals forces). Chemisorption is characterized by higher interaction energies which result in a chemical modification of  the adsorbed component. The adsorption capacity (loading) of an adsorbent for a given component is normally

represented as a function of the component’s concentration c in the gas (or liquid) for the equilibrium condition at constant temperature, known as the adsorption isotherm x ¼ f (c)T . There are a variety of approaches derived from different model assumptions for the quantitative description of adsorption isotherms (see also  Adsorption, Chap. 4,  Air). The Langmuir isotherm is based on the assumption of ideal monolayer adsorption. The BET isotherm additionally takes into account multilayer adsorption. However, their underlying assumption of a homogeneous surface structure is not applicable to adsorbents whose structure is characterized by active sites with widely differing bonding energies. In such cases, the empirical Freundlich isotherm is often useful. According to the Freundlich isotherm, the logarithmic adsorbent loading increases linearly with the partial pressure of the component to be adsorbed in the carrier gas. However, commercial adsorbents do not have a smooth surface but are highly porous solids with a very irregular and rugged inner surface. This fact is taken into account by the potential theory which forms the basis of the Dubinin isotherm. It describes the logarithm of  the adsorpt volume V , which is determined via the density of the liquid phase of the component to be adsorbed as a function of the relative saturation (log p /  ps)n. According to [12, 13], this equation is applicable to some wide-pore activated carbons with an exponent of  n ¼ 1, while an exponent of  n ¼ 2 results for a variety of activated carbons with organic vapors and certain carbon molecular sieves. At adsorption temperatures below the critical temperature of the component to be adsorbed, the adsorbent pores may fill up with liquid adsorpt. This phenomenon is known as capillary condensation and enhances the adsorption capacity of the adsorbent. Assuming cylindrical pores, capillary condensation can be quantitatively described with the aid of the Kelvin equation, the degree of pore filling being inversely proportional to the pore radius. In the liquid phase, the empirical Freundlich isotherm is a very helpful tool, as the position and gradient of this isotherm allows conclusions to be drawn regarding carbon properties in practical applications. Often, specific substances like phenol, iodine,

Carbon, 5. Activated Carbon Table 5. Typical design dataof adsorbers for adsorptivegas and water purification

Parameter

Gas purification

Water purification

Carbon particle size, mm Depth of adsorbent bed, m Mass transfer zone, m Superficial velocity, cm/s Residence time, s

3–5 0.5–1.5 0.05–0.3 10–50 1–15

0.5–2.5 2–15 0.5–5 0.03–0.4 a 1800–7200

a b

b

1–15 m/h. 0.5–2 h.

alkylbenzenesulfonate, or methylene blue are used for characterizing the adsorptive properties of activated carbons. For commercial adsorption processes not only the equilibrium value, but also the rate at which it is achieved (adsorption kinetics) is of  decisive importance. The adsorption kinetics are determined by the following series of individual steps: *

* * *

Transfer of molecules to the external surface of the adsorbent Boundary layer film diffusion Diffusion into the particle Actual adsorption step

Adsorption kinetics in the liquid phase are much slower than in gas phase; therefore adsorber design is different for liquid and gas phase applications (see Table 5).

2.5. Quality Control Methods of quality control are of great importance for both technical and commercial practice. Some tests are used in production control to check the basic properties of final or intermediate products. Special requirements are placed on acceptance tests, since they may be used in disputes between buyer and supplier. The buyer wishes to be certain of the material’s suitability for his purpose and uses tests in which the conditions simulate as closely as possible his operational conditions. For example, active carbon for sugar refineries is tested for its decolorizing efficiency on a molasses solution to determine the adsorption properties but also side effects due to the ash content and pH value. A gas-mask carbon is tested dynamically for its service time with the substances that are to be removed in practice [1, 4].

7

  Physical and Mechanical Tests [1]. Bulk  density (DIN-ISO 787 11, ASTM D2854) is defined as the mass of a unit volume of the sample in air, including both the pore system and the voids between the particles. It is expressed in kg/m3 on a dry basis. The bulk density of activated carbon depends on the shape, size, and density of the individual particles. Bulk  density data are useful for the estimation of  tank or packing volume. The particle density, otherwise known as Hg density, is defined as the mass of a unit volume of the carbon particle, including its pore system, normally expressed in g/cm3. The particle density is an important characteristic of granular carbon that is used for the determination of bed porosity or void fraction. This in turn, is necessary for the determination of numerous other properties. Under the conditions of this method, pores r  ! 7 mm or more in radius are filled and therefore do not contribute to the density of the particle. The absolute or helium density is defined as the mass of a unit volume of the solid carbon skeleton that is inaccessible to He, normally expressed in g/cm3. The pressure drop gives information about the resistance to flow of a gas through a pelleted or granular carbon layer. The pressure drop over a packed bed is adequately defined by a modified form of the semi-empirical Ergun equation as a function of the shape and size of the particles and the temperature, pressure, and superficial velocity of the gas. It is expressed in pascals per meter of carbon bed length. Air is passed through an activated carbon layer of  given length, and the pressure drop is measured as a function of gas velocity. The resistance to the flow of liquids or gases has a practical significance, since it is related to the filtration resistance of a carbon layer. The relationship between flow rate and pressure drop is shown in Figure 3. Particle size is an important property that influences the flow characteristics, adsorption kinetics and catalytic behavior of granular activated carbon layers. The grain size distribution of granular activated carbons (ASTM D2862) is determined by using standard sieves and motordriven sieving apparatus. All common methods provide for mechanical separation with standard sieves, the aperture of which is expressed in

8

Carbon, 5. Activated Carbon

millimeters or mesh. Determination of particle size by sieving is not applicable to extruded activated carbons. The fineness of powdered carbons can be determined by elutriation or by laser-beam scattering. Equipment such as the Coulter counter or sedimentation tests lead to incorrect results because the individual particles of activated carbon vary in electrical conductivity and density.  Mechanical strength is an important factor in most technical applications of granular activated carbon. Mechanical strength tests are modified to conform with the technical requirements: it differs for active carbon for gas masks, in which the granules suffer attrition, from that for solvent recovery, for which coherence of the granules is critical. The determination of  mechanical strength simulates the resistance to abrasion or attrition under practical conditions. Depending on various practical requirements, there are many different test methods using vibration, impact, rotary motion, or motion as in a fluidized bed. A variety of tests are available for the evaluation of the mechanical strength of  granular activated carbon. In these tests the change in particle size distribution or the amount of fines produced is determined. An overview of the wide range of hardness tests, which test different aspects of the mechanical strength and hence cannot be related to one another mathematically is given in the following: Ball-mill hardness:

Abrasion strength:

Impact hardness:

Ball-pan hardness (ASTM D 3802):

Crushing strength:

Impact hardness (fluidized bed):

The activated carbon is abraded for a given time in a horizontal cylinder with steel or ceramic balls under prescribed conditions The activated carbon is abraded by an iron rod in a horizontal rotating cylindrical sieve of given dimensions for a prescribed time The activated carbon particles are broken by dropping a weight onto a sample under controlled conditions The activated carbon is shaken for a given time in a pan together with a given number of steel balls of known diameter The pressure required to crush a granule of activated carbon (not applicable for broken granules) The activated carbon is pneumatically agitated for a standard time in a vertical cylinder, the top of which is equipped with an impact plate.

Attrition and The activated carbon is rotated for a abrasion resistance period of time in a cylindrical drum (ASTM D4058): having a single baffle Stirring abrasion The activated carbon is abraded by (AWWA B604): a T-shaped stirrer in a special abrasion unit Ro-Tap abrasion The activated carbon is shaken with (AWWA B604): steel balls in the testing pan of  a Ro-Tap sieve machine

Chemical and Physicochemical Tests.  Moisture content  (ASTM D2867) is determined by heating the sample in air in an oven at constant temperature to constant weight (3 h at 150  C). The oven-drying method is used when water is the only volatile material present in the activated carbon. A xylene distillation method is used when the carbon is known or suspected to be heat-sensitive or to contain water-insoluble volatile compounds as well as water.  Ash content  (ASTM D2866) is determined by ignition of the sample to constant weight in a muffle furnace (air circulation) at 650 Æ 25  C. Frequently only the water-soluble or acid-soluble part of the ash is determined. In cases where certain cations or anions have an adverse effect on the products, they can be determined by first extracting with hydrochloric acid, nitric acid, or water in a prescribed way, and then analyzing the extract. Arsenic, cyanide, and sulfide must be determined on the activated carbon itself. Volatile matter  is determined by heating the sample at 900  C for 7 min. The percentage of  volatile matter is calculated from the weight loss of the sample, corrected for moisture content.  Ignition temperature (ASTM D3466) is determined by exposing a sample of carbon to a heated air stream, the temperature of which is slowly increased until the carbon ignites. The temperature of the carbon bed and of the air entering the bed are recorded, and ignition is defined as the point at which the carbon temperature suddenly rises above the temperature of  the air entering the bed. The test provides a basis for comparing the ignition characteristics of  different carbons, or the change in ignition characteristics of the same carbon after a period of service. The self-ignition test  (IMDG Code, Class 4.2) determines whether a sample of  chemically activated carbon ignites at 140  C under specified conditions. Steam-activated carbon is considered to be non-self-ignitable.

Carbon, 5. Activated Carbon

Activated carbon bearing inorganic and chemically active groups on its surface may alter the pH of liquids to which it is added. A predictive standard test giving, a good approximation of actual conditions has been devised. The acid or alkali content is determined by boiling an aqueous suspension and measuring the pH of the filtered or decanted aqueous extract. The pH of the extract is defined as the  pH value of the activated carbon.  Adsorption Measurements. The most common method of measuring adsorption properties of activated carbon is by the determination of  the BET surface, e.g., according to DIN 66 131. However, the numerical value thus obtained has only a limited practical significance, because in a practical situation, the molecules adsorbed are usually to large to reach the inner surface of the very small pores due to the large size of their molecules, whereas determination of the nitrogen isotherm [17] or assessment of porosity by using mercury vapor [18] gives deep penetration into the pore structure. It is easier to estimate the porosity by measuring the amount of benzene or cyclohexane vapor adsorbed by activated carbon at 20  C and at a range of partial pressures. The difference between two measured values of  the isotherm represents a certain pore volume, and limiting values of pore sizes can be related to this. An example of the single-point method is the determination of the carbon tetrachloride retentivity according to ASTM [19]. Carbons for gas masks are characterized by the breakthrough times (holding times or service times), which are determined by using certain test substances such as chloropicrin. For many practical applications involving gas-phase adsorption, the total adsorption capacity is of minor interest in comparison with the adsorption capacity after regeneration. In such cases, the “working capacity” after several cycles of adsorption and desorption is quoted.

3. Production 3.1. General Aspects Nearly all carbon-containing materials can be used for the manufacture of activated carbon; e.g., wood, nut shells, fruit stones, peat,

9

charcoal, brown coal, lignite, bituminous coal, mineral oil products, and some waste materials. Cellulose and organic polymers are used for the manufacture of fiber and spherical activated carbon [2–6]. These starting materials vary considerably with regard to the extent to which they can be activated; e.g., calcined petroleum coke or high-temperature coke from coal are difficult materials, while wood charcoal is easily activated. In addition, the purity of the activated carbon produced as well as its pore size distribution is very much dependent on the starting material. There are two principal methods of activation, i.e., that which uses chemicals and that which uses gases. Chemical activation is based on the dehydrating action of certain substances, e.g., phosphoric acid [7664-38-2] or zinc chloride [7646-85-7], mostly on uncarbonized starting materials such as sawdust or peat. A temperature of 400–1000  C is usually used. After removal of the chemicals, e.g., by extraction, the porous and active carbon structure of the raw material remains. Cellulose fibers or woven materials produce products having the same physical form, but made of activated carbon. Polyacrylonitrile fibers are also used. Gas activation entails the use of gases containing combined oxygen, such as steam or carbon dioxide. At temperatures of 800– 1000  C, some of the carbonaceous starting material is decomposed, producing numerous extremely fine pores or cracks. The inner surface area of the carbon determines its degree of  activation. The yield is, therefore, dependent on the degree of activation. A high degree of activation is associated with a low yield, and this can be between 20 and 60%. Production of activated carbon can be carried out in rotary kilns, multiple hearth furnaces, or furnaces of the vertical-shaft or fluidized-bed type, each type being suitable for a particular particle size of starting material. At the present time, internally heated rotary kilns are most commonly used, since these are suitable for production of activated carbon of a large range of particle sizes from powder and granular material up to cylindrical pellets. The activation of coarse materials such as softwood and beech wood charcoal is carried out in a shaft furnace. The product is usually ground to a powder for use as a decolorizing agent. Fluidized-bed

10

Carbon, 5. Activated Carbon

furnaces are suitable for granules and cylindrical pellets. Important factors that determine the adsorption properties of an activated carbon are the pore volume, the pore size distribution, and the type of functional groups on the surface (surface oxides). The large inner surface area is mainly due to the micropores. Consequently, particular attention is paid to these during manufacture. Pore volume and pore size are affected by the type of starting material and the heating process. Thus, gas activation of coconut shell charcoal always gives a high proportion of fine pores, while the same process with softwood charcoal yields a product with open pores. Chemical activation produces carbons with extremely high proportions of meso- and micropores.

3.2. Raw Materials Wood, sawdust, peat, straw, and other cellulosecontaining materials are usually only treated by chemical activation. The direct gas activation of uncarbonized products is possible, although usually with these raw materials, a carbonization process (possibly at low temperature) is added. This has the advantage that the partly carbonized intermediate product can be screened, giving a standard particle size. Norit NV (Netherlands) carries out gas activation of peat without producing any partially carbonized intermediate material. In a further development of the classical procedure of OSTREJKO [7], activated carbon is produced directly from uncarbonized carbonaceous material by heating to 840–900  C in an atmosphere of combustion products that contains hydrocarbons but no oxygen [8]. Raw materials very suitable for gas activation are wood charcoal, nut shell charcoal, and coke from brown coal or peat. Wood charcoal [7440-44-0] is obtained by carbonization of  pieces of wood from beech, spruce, or pine in large-capacity batch retorts or continuous vertical retorts ( Charcoal). Carbonized coconut shells are used by many producers in Europe, Japan, and the Far East.  Bituminous coals are suitable for gas activation to varying extents. A useful criterion is their carbon, oxygen, and hydrogen contents. The diagram in Fig. 5 gives an approximate guide

Figure 5. Characterization of different raw materials

to these values [20]. The high-carbon materials graphite and anthracite are very difficult to activate. Bituminous coals with high oxygen and hydrogen contents can have troublesome sticking and swelling properties. In these cases, a preoxidation is usually carried out before gas activation [21, 22]. This is done with air or oxygen-containing gases at temperatures between 150 and 350  C. This causes 5–30% oxygen to be taken up. The optimum temperature is 220–250  C. This procedure is carried out as long as necessary to cause complete disappearance of the swelling and sticking properties; as much as 5 h may be required. Bituminous coal may be more easily activated after addition of mineral acids such as phosphoric acid. For this purpose, the raw material is finely ground and mixed with a few percent of  acid before being formed into shapes. This process, which was developed by Carborundum [23], is operated in the United States by Ceca. It uses a combination of chemical and gas activation. In some types of bituminous coal, the quality of the activated product is adversely affected by the high ash content; however, a small ash content that includes potassium compounds, for example, can have a catalytic effect and, when the material is preoxidized, can lead to intensive gasification of the coal. In these cases the ash content of the raw material can be

Carbon, 5. Activated Carbon

reduced by grinding and flotation. The preoxidation can be carried out on the powder or after granulation. Some companies which activate bituminous coal: the Calgon Carbon Corp. (United States), Norit (United States), and the CarboTech Aktivkohlen GmbH (Germany).  Brown coals and the related lignites offer an inexpensive and readily activated starting material for the production of activated carbons. Their relatively high ash and sulfur contents are detrimental. Therefore, methods have been developed for removing ash from brown coal before coking, e.g., by treatment with an oil– water mixture. The ash goes into the aqueous phase, while the coal remains in the oil phase. By this means the ash content can be reduced by 80–90%. Lignite is activated in high tonnages by Norit in the United States according to a process developed by ICI United States [24]. Many publications and patents describe the production of activated carbon from mineral oil  products . Petroleum sludges, fly ash, old tires, domestic refuse, and sewage sludge have also been proposed as raw materials for activated carbon, but none of these products have attained commercial importance [25, 26]

3.3. Activating Furnaces Shaft furnaces originally consisted of simple vertical chambers with smooth walls made of  refractory bricks. Heating is external. As the process was developed, the mixing of the feed materials was improved and, consequently, so was the reaction with the activating gas. This was done by installing ceramic attachments in the form of gratings or replaceable louvers which can control the direction and velocity of  the gas stream within the furnace. The reaction gases (hydrogen and carbon monoxide) can be removed at various levels. The temperature of  the furnace, usually 5–8 m high, can be controlled by means of a number of burners or afterburners (Fig. 6) [27]. Shaft furnaces can be used for the reactivation of exhausted activated carbon [28].   Rotary kilns are the most commonly used activating furnaces. Due to the length of the kiln and the high temperatures necessary for gas activation, direct heating is the only feasible method when the materials of construction of 

11

Figure 6. Shaft furnace

the kiln are considered. To be able to control gas composition and temperature throughout the entire length of the kiln, several burners and gas supply lines are distributed along and around the kiln casing, e.g., in the American Norit furnace. Figure 7 shows an arrangement including lifters to give improved mixing of the feed material. By means of a variable steam injection rate, the water vapor content and, therefore, the activation rate can be further controlled.  Multiple-hearth furnaces (Fig. 8) with rotating arms and stationary floors on each stage are used by several firms such as Calgon in the United States and Belgium. This type of furnace is also operated for the purpose of reactivation, e.g., by Windhoek Municipality (Republic of  South-West Africa). Fluidized-bed furnaces offer the advantage of extremely intensive heat and mass transfer. This means not only that the activating gases are quickly brought into contact with the raw material, but also that the waste gases are just as quickly removed. Furnaces have been developed which are operated continuously (Fig. 9)

12

Carbon, 5. Activated Carbon

Figure 7. Rotary kiln for steam-activation process a) Steam; b) Gas; c) Air; d) Burner; e) Brick lining; f) Lifters

and in which several fluidized beds are run in series. The activating gases may be introduced into the spaces between the fluidized layers and also into the circulatory system. Thus, it is

possible to achieve a preoxidation in the first stage with oxygen-containing gas and then to carry out the actual activation with oxygen-free gases in a second stage. The activation in the fluidized bed is so intensive that usually only fragile products with poor resistance to abrasion

Figure 8. Multiple hearth furnace a) Raw material silo; b) Inlet; c) Burner; d) Off-gas suction; e) Outlet for activated carbon; f) Dust collector; g) Off-gas stack 

Figure 9. Fluidized-bed furnace a) Raw material silo; b) Inlet; c) Combustion chamber (indirect heating); d) Burner; e) Gas distribution plate; f) Outlet for activated carbon; g) Heat exchanger

Carbon, 5. Activated Carbon

are obtained, which are processed to give decolorizing carbon in powder form. By maintaining certain conditions, in particular by keeping the water vapor content of the gas below 0.6 kg/m3 and by maintaining a neutral or slightly reducing atmosphere, it is possible to produce abrasion-resistant granules.

3.4. Methods of Activation 3.4.1. Chemical Activation  Zinc Chloride Process. In the classical zinc chloride process for the chemical activation of  carbonaceous materials, 0.4–5.0 parts of zinc chloride as a concentrated solution are mixed with 1 part peat or sawdust. The mixture is then dried and heated to 600–700  C in a rotary kiln. The product is washed with acid and water, and the zinc salts are recovered. In some cases, chemical activation is followed by steam activation to obtain additional fine pores. In spite of  the efficiency and simplicity of the process, it is in decline because of the problems of environmental contamination with zinc compounds. Phosphoric acid [7664-38-2] can be used to treat either uncarbonized or carbonized raw materials, and the process is operated by Ceca (France), Hooker (Mexico), and Norit (United Kingdom and United States). Finely ground raw material such as sawdust is mixed with a phosphoric acid solution, forming a pulp. This is dried and heated to 400–600  C in a furnace such as a rotary kiln. The phosphoric acid is then extracted, sometimes after neutralizing it to give phosphate salts, and the material is dried, giving an activated carbon which usually has finer pores than the zinc chloride product. Activation with a combination of phosphoric acid and steam is also possible. As with zinc chloride activation, a highly active decolorizing carbon is obtained by a rapid process in high yield and at a relatively low reaction temperature. However, the cost of  recovering the activating chemicals is high. Activation by phosphoric acid has become more popular and there is no doubt that improved methods of phosphoric acid recovery have contributed to this. These innovations have hardly been reported in the literature; the know-how is not divulged by the producers.

13

Other Chemicals. In the literature, many chemicals have been proposed for the activation of carbonaceous raw materials, but none of them attained industrial importance.

3.4.2. Gas Activation

In gas activation, carbonaceous material is treated at elevated temperatures with suitable gases, the most common being steam, carbon dioxide, and mixtures thereof. Experiments using graphite have established the reaction velocities: steam has been shown to be 8 times as reactive as carbon dioxide. Both gases behave as mild oxidizing agents at 800–1000  C, there being several simultaneous reactions: H2 OþC  COþH2 DH ¼ þ117 kJ 2H2 OþC  CO2 þ2H2 DH ¼ þ75 kJ CO2 þC  2 CO DH ¼ þ159 kJ

Due to the endothermic character of these reactions, the carbon particles must be brought into intimate contact with the activating gas. This must be hotter than the required reaction temperature; otherwise, the necessary heat energy will not be provided, or only with difficulty. Below 800  C, the reaction velocity is so seriously reduced that the activation process ceases for all practical purposes. A useful improvement to the heat supply can be obtained by combustion of gases produced during activation: COþ0:5 O2  CO2 DH ¼ À285 kJ H2 þ0:5 O2 H2 O DH ¼ À238 kJ

Modern furnace construction takes advantage of this fact by introducing oxygen and air at suitable points, which at the same time has the effect of regenerating the activating gases. Thus, the best way of dealing with the carbon monoxide and hydrogen that are produced is by burning them off in the reactor itself. This is necessary for the additional reason that these gases reduce the velocity of activation, carbon monoxide to a noticeable extent and hydrogen very markedly. The reaction mechanism for the gas activation of carbon with steam or carbon dioxide is characterized by an initial adsorption of 

14

Carbon, 5. Activated Carbon

these gases with subsequent oxidation of the carbon surface as the rate-determining step: CþCO2  CðOÞþCO CðOÞ  CO

where C(O) signifies surface oxide. The retarding action of the carbon monoxide and hydrogen can be attributed to the formation of C(CO) and C(H) surface complexes, the latter in particular being much more stable than the C(O) surface complex. Thus, the active sites which could adsorb oxygen are blocked by hydrogen. Oxygen or air are unsuitable as activating gases. In a mixture with steam or inert gas, small amounts of oxygen lead to activated material with very large pores. Oxygen reacts with carbon about 100 times as fast as carbon dioxide. This reaction velocity is even further increased by potassium salts, so that potassium-containing raw materials react so vigorously when oxygen is present in the gas that an uncontrolled combustion takes place without producing activation. The chemical condition of the carbon surface, especially the presence of larger or smaller amounts of carbonyl and carboxyl groups, can determine the adsorption properties of the activated carbon and, very importantly, its properties as a catalyst. Acidic surface oxides are formed by heating activated carbon in air or oxygen for a short period below the ignition temperature. If carbon is first heated to 1000  C and then allowed to react with air at room temperature, basic surface oxide groups are formed, although the quantity of these is at most much less than the quantity of  acidic groups. It has long been known that the gas activation of carbonaceous materials is accelerated by small amounts of various compounds, e.g., salts of alkali and alkaline earth metals, almost all chlorides, sulfates, acetates, and carbonates, as well as most acids and hydroxides. The most important catalysts used industrially are caustic potash and potassium carbonate. Amounts between 0.1% and 5% are used. The activation accelerators can be used in solid form mixed with the finely powdered carbonaceous substances or added as solutions, sometimes followed by molding into shapes and low-temperature carbonization.

If bituminous coal is activated with addition of alkali metal salts, the gas mixtures containing carbon dioxide must be used for the activation process in preference to pure steam. In addition to these accelerators, the patent literature also refers to compounds of iron, manganese, and aluminum. The theory of catalytically accelerated activation is described in [29] and [30].

3.5. Granular and Pelletized Carbons For a number of applications in liquid and gas purification, the activated carbon must be provided in the form of grains or granules; therefore, the raw material used to produce it must be similarly shaped, e.g., wood charcoal or coconut shell charcoal. Alternatively, the raw material, such as coal, can be pulverized, briquetted by using a binder, and finally carbonized. Before activation, the material is broken down to the required particle size. For applications requiring a carbon with a high mechanical strength, it is often an advantage to use carbon which has been specially preshaped, i.e., pelletized. In some cases, the hardness is dependent not only on the raw material and binder, but also on the degree of activation and porosity. Highly activated carbons have a high pore volume and, therefore, low density and strength. Similarly, products with open pores are not as hard as those with fine pores. Carbons in pellet form are produced from finely powdered raw material as follows. The powder is first mixed with a binder in a heated paste mixer to give a flowable mass. This material is then extruded to form strands ! 1 mm in cross section, or may be formed discontinuously in a cylinder press (Fig. 10). These strands, sometimes after drying, are then broken into short lengths, the length of each piece being approximately equal to its diameter. These are then either chemically activated or carbonized at 400–500  C and finally gas-activated. Suitable binders are coal tar, wood tar, lignosulfonic acids, or mixtures of phenols and aldehydes or their condensation products. Bases such as caustic soda or chalk neutralize the acidic groups of  the tar and improve gas activation. A simplified flow sheet of the CarboTech process is shown in Figure 11. The specially

Carbon, 5. Activated Carbon

15

aqueous medium, it is necessary to disintegrate them. There are a number of patents describing the production of microporous carbon spheres from pitch. The process involves several stages: melting, dispersing, oxidizing with air to render the material insoluble, and finally, activation by steam.

3.6. Carbon Molecular Sieves Figure 10. Flow sheet for production of pelletized activated carbon a) Crusher; b) Mill; c) Kneader; d) Extrusion; e) Drying; f) Carbonization; g) Activation; h) Screening; i) Packing

de-ashed hard coal feed is finely ground, and partially oxidized with air. Then the coal dust is mixed with a binder. The binder is added to produce a plastic coal/binder mixture that can be used to form shaped extrudates of the desired diameter. The extrudates are then carbonized to activated coke in a rotary kiln at 900–1000  C. In a final activation step, the carbon skeleton of  the activated coke is partially gasified by steam activation in a multistage fluidized-bed furnace. The production of pelletized activated carbon from material which is already in an activated state, such as powdered decolorizing carbon, is at the present time of little commercial importance. In the beverage and food industries, products of this type have found increased application due to their dust-free nature. To make the fullest use of the available adsorption capacity of agglomerates of this type in an

Figure 11. Production steps of formed activated carbon

Carbon molecular sieves for air separation are produced from coal, coconut shells, or resins. For example, bituminous coal is ground to a fine dust and oxidized in air at a temperature below the ignition temperature. The oxidized coal is mixed with a binder and shaped into pellets with diameter of 2.5 mm. The pellets are carbonized in a special rotary kiln. The final step is treatment with hydrocarbon under cracking conditions to deposit carbon exactly on pore openings [31]. The pore diameter is then smaller than that of the initial material, and the diameter of the “bottle necks“ are on the same range as those of nitrogen and oxygen molecules. Oxygen molecules can penetrate much quicker than the nitrogen molecule into the pores. Therefore, the most of the nitrogen is recovered, while almost all of the oxygen is adsorbed.

3.7. Further Treatment Many firms produce low-ash varieties of activated carbon by removal of various impurities

16

Carbon, 5. Activated Carbon

by washing with water or acids such as hydrochloric or nitric acid. If the activated carbon is to be used for the production of fine chemicals or pharmaceutical preparations, a particularly thorough washing procedure is required, and activated carbons for use as catalysts or catalyst support require similar treatment. Basic constituents and accelerators such as caustic potash may be neutralized by acid or washed out with water. Active carbons with fine pores, made by steam activation at 800  C, can be further activated by air at 500–600  C in the presence of  alkali, giving improved decolorizing ability. However, this two-stage activation process has no known practical importance. The most important result of a postoxidation of activated carbon by nitrous gases, in particular nitrogen dioxide, is the formation of additional surface oxides rather than any further activation. The opposite effect, i.e., the removal of chemically bound oxygen, is possible by treating activated carbon with hydrogen at 200–500  C. Lowering of sulfur content can be achieved by the action of  steam and hydrogen. In the literature a method is described of producing activated carbon with a low iron content. This involves converting the iron into volatile compounds by treating the carbon while it is still hot with suitable gases or vapors such as halogens, halogen compounds, or carbon monoxide.

activated carbon is sprayed in a rotary kiln or in a fluidized bed under defined conditions. The impregnated wet activated carbon must be dried in, for example, a rotary kiln or fluidized-bed drier. After the drying step, most impregnated activated carbons can be used industrially. In some applications the impregnating agents are present in the form of hydroxides, carbonates, chromates, or nitrates and must be subjected to thermal aftertreatment at higher temperatures (150–400  C) to decompose the anions. Depending on the application, various activated carbons (pellets, granules, powders) are impregnated with suitable organic or inorganic chemicals. Homogeneous distribution of the impregnating agents on the internal surface of an activated carbon is important. Furthermore, blocking of the micropores and macropores must be avoided so that the impregnating agent remains accessible for the reactants.

4. Applications Since the range of applications for carbonaceous adsorbents in the gas and liquid phases is very broad only a small selection can be dealt with here. The major applications are in water treatment, gas purification, food processing, gold recovery, and solvent recovery (Fig. 12).

4.1. Gas-Phase Applications 3.8. Impregnation For cost-effective removal of certain impurities contained in gases (e.g., hydrogen sulfide, mercury, and ammonia), the adsorption capacity and removal rate must be substantially increased by impregnation of the activated carbon with suitable chemicals. When these chemicals are deposited on the internal surface of the activated carbon, the removal mechanism also changes. The impurities are no longer removed by adsorption but by chemisorption [32]. For the manufacture of impregnated activated carbon, an activated carbon of suitable quality for the particular application is impregnated with solutions of salts or other chemicals which, after drying or other aftertreatment steps, remain on the internal surface of the activated carbon. As well as soaking impregnation, spray impregnation can be used. In that case the

The majority of gas- and vapor-phase applications of activated carbon are in process gas purification, air purification, catalysis, flue gas

Figure 12. Gas- and liquid-phase applications of carbonaceous adsorbents

Carbon, 5. Activated Carbon

Figure 13. Gas-phase applications of adsorbents

carbonaceous

purification, solvent recovery, and automotive emission control, and personal protection (Fig. 13).

4.1.1. Solvent Recovery

Activated carbon can be used advantageously for the removal of organic vapors from gases, its performance being good even at very low partial pressures. This may be deduced from the linear adsorption isotherm plotted in Figure 14. For example, a pelletized activated carbon is able to take up ca. 18% of its mass of toluene from a current of air that contains only 0.11 g of toluene per m3. This represents an enrichment factor of over 400 000 : 1. This ability to concentrate a substance is particularly useful for producing extremely pure gases, as well as for protecting the environment [2, 33–37]. When activated carbon is used for the recovery of solvents, which usually occurs at concentrations between 1 and 20 g/m 3, efficiencies of  > 90% are sought. The cost of regeneration, whether by steam, hot gas, or electrical heating, is usually small in comparison to the value of the

17

Figure 14. Linear adsorption isotherm for toluene

recovered solvent. The charge of activated carbon retains its effectiveness for a long time if the regeneration is carried out at a sufficiently high temperature. However, fine material which is formed by attrition and thermal or chemical stress must be removed and replaced from time to time. A survey of some areas of application of  solvent recovery is given in Table 6. In industry, the solvent recovery is carried out in vertical or horizontal adsorbers, and by arranging these in parallel a continuous operation is possible by changing over from one to another (Fig. 15) [9]. The gas flow is usually in an upward direction, whereas the desorption with hot steam is in the opposite direction. In Figure 16 a typical temperature–time graph for a cycle is given. Usually, fixed-bed adsorbers have a bed thickness of 0.8–2 m. A more recent development is the use of fluidized-bed reactors with finely granulated activated carbon [37]. Another development is the use of activated carbon fibers. Both methods are, however, rarely used. Practical solvent recovery systems use gas flow rates of 0.2–0.5 m/s. Lower flow rates would lead to better utilization of the adsorption

Table 6. Solvent recovery

Industry

Typical solvents

Plastic film and foil Printing Metal degreasing Rubber Viscose and rayon Dry cleaning Synthetic leather and fibers Adhesives

ether, acetone, methyl ethyl ketone, alcohols, methylene chloride, tetrahydrofuran, cyclohexanone toluene, petroleum spirit, trichloroethene, n-hexane trichloroethane, trichloroethene, tetrachloroethene petroleum spirit, benzene, toluene carbon disulfide tetrachloroethene, fluorochloro hydrocarbons alcohol, acetone, hexane, toluene, esters, dimethylformamide petroleum spirit, hexane, toluene

18

Carbon, 5. Activated Carbon Solvent adsorbed per cycle Steam/solvent ratio Energy Cooling water Activated carbon

10–25 wt% (2–5):1 50–600 kWh/t solvent 30–100 m3 /t solvent 0.5–1 kg/t solvent

Recovery units have at least two, but more usually three or four adsorbers which pass successively through the stages of the operation cycle. While adsorption takes place in one or more of them, desorption, drying, and cooling are carried out in the others.

4.1.2. Process-Gas and Air Purification

Figure 15. Flow sheet of a solvent recovery unit a) a1) Adsorber 1; a 2) Adsorber 2; b) Exhaust air; c) Blast; d) Desorption; e) Condenser; f) Cooler; g) Separator

Figure 16. Temperature diagram for the carbon bed of a solvent recovery unit a) Upper part; b) Middle; c) Lower part

capacity of the carbon, but there is a danger that the heat of adsorption is not carried away, and therefore overheating and even ignition of the carbon charge can occur. Typical operating data for solvent recovery plants and design ranges are given in the following: Air velocity Air temperature Bed height Steam velocity Adsorption Drying (hot air) Cooling (cold air) Solvent concentration

0.2–0.4 m/s 20–40  C 0.8–1.5 m 0.1–0.2 m/s Time cycle per adsorber 2–6 h 0.2–0.5 h 0.2–0.5 h 1–10 g/cm 3

Many gas purification processes use activated carbon, e.g., for production of pure gases in the chemical industry, in protection against poison gas, in air conditioning, for removal of oil from compressed air, and in purification of waste air. Small traces of unwanted gases or vapors are adsorbed onto activated carbon which is often improved by being specially impregnated, in which case regeneration with recovery of the extracted materials is impossible. Activated carbon is also used to remove resin-forming and other hydrocarbons from gases before passing them over sensitive molecular sieves or catalysts [2, 9, 31–36]. Small amounts of hydrogen sulfide can be converted to elemental sulfur in the presence of  oxygen by means of activated carbon which has been impregnated with potassium iodide [36]. This reaction is used in the viscose industry for purification of waste air. In this case, it is combined with carbon disulfide recovery and is known as the Sulfosorbon process [36, 38]. The Sulfren process uses sulfur dioxide as an oxidizing agent. Sulfur compounds such as carbon disulfide, carbonyl sulfide, and organic thiols are removed from moist gases containing excess oxygen at high temperature on alkaline carbon. This is the Desorex process. The adsorbate is hydrolyzed and oxidized to sulfate, which can be washed out. In air-conditioning installations, activated carbon is used for the purification of air drawn in from outside, e.g., in airports, near chemical plants, or for environmentally controlled rooms of hospitals or museums. By the use of activated carbon, the amount of necessary cold outside air

Carbon, 5. Activated Carbon

can be reduced. Because of the large quantities of air to be handled for such large spaces, a small resistance to flow is required, and therefore, thin layers of carbon are used. Often exchangeable cartridge filters are used, and complete filter elements made of specially shaped activated carbon bodies are available. Composite materials have been produced, in which powdered activated carbon is bonded onto polyurethane foam or some other suitable carrier. For the removal of some substances, impregnated activated carbons like those in industrial respirators are used. In nuclear power installations, activated carbon impregnated with iodine compounds is used to remove radioactive iodine compounds from the air exhausted to the atmosphere. This takes place by isotope exchange. In some countries impregnation with the base tetraethylenediamine (TEDA) is used for the same purpose; in this case the iodine compounds are removed by salt formation. Activated carbon is also used in off-gas delay beds. Deep beds of activated carbon with very fine pores adsorb radioactive gases such as krypton and xenon long enough for the isotopes to decay to safe levels of radioactivity, after which they can be released into the atmosphere. Since World War I, activated carbon filters have been used by the armed forces for respirators, and this use has since been extended to protection against hazardous gases in industry. For both applications, virtually complete removal of impurities is necessary, and therefore, only fine-grained activated carbon can be used so as to make a filter as compact as possible. Table 7 shows how the service time depends on the grain size. For respirators, impregnation of  activated carbon with chromium and copper salts is a well-proven technology; due to the toxicity of chromium compounds the latest impregnations are based on copper and molybdenum salts. These substances have a strong oxidizing action and high reactivity toward chlorine, hydrogen cyanide, and their derivatives.

Table 7. Effect of grain size on the service time of a respirator fitted with an activated carbon filter (test gas: chloropicrin in moist air)

Diameter of carbon granules Service time

mm min

0.8 75

1.2 60

1.6 35

2.5 18

4.0 5

19

Other substances used for impregnation are caustic potash (acidic gases), zinc salts (ammonia and hydrogen sulfide), and iodine compounds (mercury vapor). Activated carbon and impregnated activated carbon is also used for many odor emission control applications: wood chip drying (pinene, terpene), plastic processing (styrene, benzene, etc.), home application (kitchen hood, refrigerator). Apart from physisorption, chemisorption on impregnated activated carbon (Section 4.3) can be applied, especially for H2S and mercaptan removal. In cigarette filters and in attachments for tobacco pipes, activated carbon is used to reduce the nicotine and tar content of the smoke.

4.1.3. Gas Separation

In 1960 the first pressure-swing adsorption (PSA) plants for gas drying, gas purification and gas separation were built. All PSA processes have in common that adsorption is operated at a higher and desorption at a lower total pressure. In the majority of processes the adsorption pressure is markedly higher than atmospheric pressure. For desorption the pressure is either reduced to atmospheric pressure or vacuum is applied to lower the pressure below atmospheric pressure. In some processes, desorption is boosted by a flushing cycle. The pressure-swing adsorption cycle comprises the following steps: * * * *

Adsorption at higher pressure Desorption by pressure reduction Flushing with product gas Pressure build-up with raw gas or product gas to adsorption pressure

Intervals between these individual steps are quite short (30 s to a few minutes) and depend on the separation process and plant design. Several types of product gases can be obtained during the absorption or desorption step, see Table 8 [31, 39]. For gas separation, carbon molecular sieves or activated carbons with extremely fine pores and molecular sieve properties are used. From a mixture of hydrogen, carbon monoxide, and

20

Carbon, 5. Activated Carbon

Table 8. Selected application fileds for pressure-swing processes for gas separation b

Separation problem

Adsorbent

Production phase a

Gas drying Hydrogen from coke-oven or reformer gas Helium from diving gases Nitrogen from air

Al2O3, SiO2, ZMS CMS, ZMS, Al 2O3

A A

CMS, ZMS, Al 2O3 CMS ZMS Oxygen from air CMS ZMS Methane from biogas CMS Carbon dioxide from exhaust gas CMS, ZMS

A A D D A A D

a

A ¼ adsorption phase, D ¼ desorption phase. Al2O3 ¼ aluminium oxide; CMS ¼ carbon molecular sieves; SiO2 ¼ silica; ZMS ¼ zeolitic molecular sieves. b

methane, it is possible to produce highly pure hydrogen by adsorption of CO and CH 4 under pressure (1.5–4.0 MPa). If pure hydrogen is then passed through in the opposite direction at lower pressure (0.2–0.5 MPa), the adsorbed substances are again desorbed. Since the holding time of an adsorber which operates in this PSA mode is in the range of 30 s to a few minutes, a large number of adsorbers (8–16) are used in rotation, so that pressure changes can be minimized between the adsorbers and compression energy can be saved. The method can be used to separate oxygen and nitrogen, as well as for production of synthetic natural gas from biogas (biologically produced methane).

4.1.4. Gasoline Vapor Adsorption

Gasoline vapors evaporate during manufacture, distribution, refueling, and running of cars and enter the environment. Vapor recovery units are installed at tank farms and distribution terminals of refineries. These waste air streams, saturated with organic vapors, are often cleaned by combined processes [40]: * *

*

Absorption and pressure-swing adsorption Membrane permeation and pressure-swing adsorption Condensation and adsorption

To avoid pollution of the environment by gasoline vapor from motor vehicles, installation of an activated carbon filter in the ventilation

port of the gasoline tank is required in the United States, the EU, and Japan. When a car is left standing in the hot sun, for example, the gasoline that evaporates from the tank is adsorbed by the activated carbon and then desorbed again when the vehicle runs and fresh air for the engine is drawn through the carbon cartridge.

4.1.5. Flue Gas Cleaning

For SO2 and NO x removal from flue gases from power plants and waste incineration plants many activated carbon/activated coke processes have been developed. The adsorption capacity of activated carbon and activated coke for sulfur dioxide from flue gas is only a few percent by weight. Therefore, processes are based on the ability of activated coke to oxidize sulfur dioxide in the presence of oxygen, forming sulfuric acid. 2 SO2 þO2 þ2 H2 O  2 H2 SO4

The spent activated coke can be regenerated thermally at 400–500  C in a desorber; the carbon of the activated carbon skeleton is used as a reactant. 2 H2 SO4 þC  2 SO2 þCO2 þ2 H2 O

This carbon consumption results in a larger inner surface area and higher catalytic activity, converting the activated coke to activated carbon. The SO2-rich gas can be processed to elemental sulfur or sulfuric acid. The catalytic properties of activated carbon are used for catalytic NO reduction by addition of gaseous ammonia. 4 NOþ4 NH3 þO2  4 N2 þ6 H2 O

Activated coke processes for simultaneous SO2 and NO x removal are the Mitsui Mining Process, the Sumitomo Heavy Process, and the Uhde/Bergbau-Forschung/Mitsui Process [41]. In the Sulfacid process, which is designed for SO2 removal, regeneration by water extraction continuously yields dilute sulfuric acid [36]. In waste incineration plants, powdered activated carbon is injected into the flue gas stream through an atomizer. The loaded adsorbent is collected in a fabric filter; further pollutant removal takes place in the filter cake deposited on the fabric [42].

Carbon, 5. Activated Carbon

4.2. Liquid-Phase Applications Liquid-phase applications are estimated to account for over two-thirds of world activated carbon consumption; both granular and powdered activated carbons are in use [2, 4, 5, 34, 44]. There are many processes available for treatment of liquids and solutions with activated carbon which have found wide industrial application. In the batch contact unit operation, powdered activated carbon is added to the liquid being treated, either directly or (more usually) as a previously prepared suspension. The temperature is raised to reduce the viscosity and hence also the diffusion time. Equilibrium is normally reached after 15–30 min, after which the mixture is filtered, usually after adding a filtration aid such as diatomite. In continuous-layer filtration, the liquid is pumped through a prepared bed of powdered carbon, which is usually made more permeable by the addition of diatomite. Due to the short contact time, the purification process is usually incomplete, and because of the limited amount of carbon present in the bed, the operating life is short. The method is used for liquids with only low levels of impurities or for purification of  working liquids which can be pumped so as to bypass the filter, e.g., in electroplating, in swimming baths, or in dry cleaning machines. Percolation through granular carbon is used particularly in the field of water purification and decolorization. This continuous process has the advantage that large charges of carbon can be employed, particularly when several filters are arranged in series. Depending on the concentration and viscosity of the solution, the contact time in the percolation process can range from 10 min to more than 1 h. In some applications catalytic side reactions such as oxidation may occur, or changes of pH can be brought about by the ash content of the carbon. Activated carbon and activated coke are often used in combination with other filter materials in multilayer filters for wastewater and drinking water treatment [45].

21

In drinking water , activated carbon is used to remove unpleasant odors and tastes and reduce the concentration of compounds constituting a health hazard (pesticides, chlorinated hydrocarbons, etc.). Powdered activated carbon is often used to solve temporary pollution problems and is added to the water as a slurry at the same time or just before adding of flocculant. After a suitable contact time, the powdered activated carbon is removed with the flocculant by sedimentation. When granular activated carbon is used to purify drinking water, after a rapid gravity sand filtration (removal of suspended solids) the percolation process is used almost exclusively, and in addition to the adsorptive purification, the catalytic decomposition of the oxidizing agents chlorine, chlorine dioxide, and ozone plays an important role. The filtration velocities are between 5 and 20 m/h, which for a layer depth of 2–4 m corresponds to a contact time of  6–48 min. The backwashing velocities are 40–50 m/h. After a service life of 6–24 months, the spent carbon is reactivated in a on-site reactivation plant or in the supplier’s reactivation plant. Figure 17 compares the different treatment steps of drinking water with powdered (PAC) and granular activated carbon (GAC).

4.2.1. Water Treatment

Water treatment (35% of world consumption) can be divided into drinking water, industrial and municipal wastewater, and groundwater.

Figure 17. Drinking-water treatment with powdered and granular activated carbon

22

Carbon, 5. Activated Carbon

The complex composition of  industrial wastewaters and dumpsite leachates mostly requires a combination of various purification methods to obtain the required purity economically. A combination of activated carbon and biological purification is suitable in many cases. Powdered activated carbon (PAC) is added to aerobic or anaerobic biological treatment plants to adsorb toxic contaminants and stabilize the biological activity. There is then no impairment of the biological activity, and the impurity-loaded carbon is removed along with the microorganisms and usually incinerated. PAC is also used independently in treatment plants [46]. Granular activated carbon filter beds are often used as a tertiary treatment after conventional secondary biological treatment. Industrial wastewaters may exhibit several times higher concentrations of organic substances than drinking water. Even though these differences in concentration are accommodated in plant design (contact time 0.5–4 h, filtration velocities of 2–6 m/h), the adsorption lifetime of the activated carbon filters is reduced from several months to few days. Accordingly, cost-effective use of activated carbon adsorbers only became possible with the development of processes for activated carbon regeneration in the 1980s.   Dump-site leachates are often purified by biological treatment followed by a filtration or membrane permeation process and an activated carbon step. The main purpose of adsorption on activated carbon is removal of organic halogen compounds.

4.2.2. Micellaneous Liquid-Phase Applications

Both powdered and granular activated carbon made from a variety of raw materials are used in   food and beverage processing. Wood-based chemically activated carbons are preferable for the removal of large color bodies and other high molecular mass impurities. Peat- and coalbased steam-activated carbons are used for decolorization and removal of unpleasant tastes, odors, and other low and medium molecular mass impurities. Microporous coconut-shell activated carbons are less efficient in decolorization. Powdered grades are used in batch processes which already require a filtration stage

and where the dosage needs to be varied according to different process conditions. The powdered activated carbon is mixed with the liquid to be purified. When the impurities have been adsorbed, the carbon is removed from the solution by filtration or sedimentation, and is then discharged. Granular activated carbons are used where high volumes of liquids of a consistent quality or grade are continuously processed in large quantities. Fixed-bed and moving bed adsorbers are used, and the spent activated carbon can be reactivated on site or by the carbon supplier. Some products which are purified with activated carbon follow: Alcoholic beverages

Soft drinks

Sugar and sweeteners

Decaffeinated coffee

Edible oils

Flavorings

Chemicals and pharmaceutical products

Beer Wine Vodka White rum Whisky Dechlorination and removal of taste and organic matter from water Decolorisation of sugar and fruit juices Cane sugar Beet sugar Polysaccharides Glucose, lactose, maltose, fructose corn syrups Xylitol Aspartame Caffeine removal by recirculating CO 2 or hot water and adsorption on activated carbon Coconut oil Palm oil Fish oil Soybean oil Olive oil Yeast extract Monosodium glutamate Hydrolysed vegetable Proteins Paraffins Waxes Phosphoric acid Sodium hydroxide Citric acid Gelatin Pectin Quinine Insulin Antibiotics Sulfonamides

Activated carbon processes have been developed gold recovery from low-grade ores. In the

Carbon, 5. Activated Carbon

carbon-in-pulp process (CIP) a suspension of  the ore is treated with cyanide to produce gold cyanide, which is then adsorbed onto granular activated carbon ( Gold, Gold Alloys, and Gold Compounds, Section 4.3. By using multistage countercurrent adsorption, the gold cyanide complex is concentrated on the activated carbon. The carbon is then separated by mechanical sieving and subjected to a elution process under slight pressure. Gold is recovered from the activated carbon by elutionfollowed by electrowinning. After thermal reactivation the carbon can be re-used in the process. Activated coconutcarbon or extruded granularcarbon,with high attrition and abrasion resistance, are used. One of the oldest uses is in medicine for the adsorption of harmful bacteria and their metabolic products in the gastrointestinal tract. For blood dialysis treatment in cases of kidney and liver diseases or poisoning, granular activated carbon with a semipermeable coating is used.

4.3. Impregnated Activated Carbon Impregnated activated carbon is predominantly used in the following applications:

* * *

23

Gas purification Civil and military gas protection Catalysis

For these applications the manufacturers offer various qualities of impregnated activated carbon. Table 9 lists frequently used impregnated activated carbons. A given impregnating agent is frequently used for various purification tasks [32]. Potassium iodide promotes the action of  activated carbon as an oxidation catalyst and thus allows catalytic oxidation of hydrogen sulfide to sulfur or of phosphine to phosphoric acid. The same impregnation technique is used to extract radioactive methyl iodide and other gaseous compounds arising in nuclear installations. Hydrogen sulfide and formaldehyde can be oxidized to nonhazardous substances with the aid of activated carbon impregnated with manganese dioxide; at high temperatures the formaldehyde is oxidized not just to formic acid, but directly through to carbon dioxide. Impregnation with iron salts and conversion to iron(III) oxide enables the removal of divalent sulfur compounds from gas mixtures low in oxygen. About 1.5 times the stoichiometric amount of 

Table 9. Commercial grades of impregnated activated carbon [32]

Impregnation Chemicals

Quantity, wt%

Activated carbon a

Examples for applications

Sulfuric acid Phosphoric acid Potassium carbonate

2–25 10–30 10–20

F 1–4 mm Ø F 1–4 mm Ø F 1–4 mm Ø

Iron oxide Potassium iodide

10 1–5

F 1–4 mm Ø F 1–4 mm Ø

Triethylenediamine

2–5

Sulfur Potassium permanganate Manganese IV oxide Silver

10–20 5

F 1–2 mm Ø G 6–16 mesh F 1–4 mm Ø, G F 3 þ 4 mm Ø G 6–16 mesh F 3 þ 4 mm Ø G 8–30 mesh

ammonia, amine, mercury ammonia, amine acid gases (HCl, HF, SO 2, H2S, NO2), carbon disulfide H2S, thiols, COS H2S, PH3, Hg, AsH3, radioactive gases/radioactive methyl iodide radioactive gases/radioactive methyl iodide

Zinc oxide Chromium–copper–silver molybdenum salts

10 10–20

Mercury (II) chloride Zinc acetate Noble metals (palladium, platinum)

10–15 15–25 0.5–1.0 0.5

a

0.1–3 0.05–0.4

F 1–4 mm Ø F 0.8–3 mm Ø G 12–30 mesh G 6–16 mesh F 3 þ 4 mm Ø f 3 þ 4 mm Ø F, G, P G 2–5 mm

mercury H2S from oxygen-lacking gases aldehyde F: phosphine, arsine G: domestic drinking water filters (oligodynamic effect) hydrogen cyanide civil and military gas protection phosgene, chlorine, arsine chloropicrin, sarin, and other nerve gases vinyl chloride synthesis, vinyl fluoride synthesis vinyl acetate synthesis organic synthesis, hydrogenation purification of terephthalic acid

F ¼ pelletized activated carbon, G ¼ granulated activated carbon, P ¼ powdered activated carbon, Ø ¼ pellet diameter.

24

Carbon, 5. Activated Carbon

oxygen is required to oxidize these compounds [36]. Activated carbon treated with elemental sulfur is used to eliminate mercury vapor from natural gas, hydrogen, andother gases [43]. Silver-impregnated activated carbon is used for purification of drinking water due to its oligodynamic effect.

4.4. Catalysts and Catalyst Supports In addition to its adsorption properties, activated carbon has catalytic properties which allow commercial use as catalyst and catalyst support. The catalytic action of activated carbon is due to the crystalline structure of the skeleton, which consists of a mixture of amorphous and graphitic carbon. On the rims of the layers, there are many unsaturated edges and ridges which act as lattice vacancies. On the internal activated carbon surface, there are surface oxides, which have been identified as carbonyl groups, lactone groups, phenolic hydroxyl groups, and carboxyl groups [6, 11]. These surface oxides can participate in redox reactions, and they are a reason for the effectiveness of activated carbon catalysts in oxidation reactions. The type and quantity of surface oxides play a part in such oxidation reactions as the conversion of sulfur dioxide to sulfur trioxide or when activated carbon is incorporated into dry cells in conjunction with manganese dioxide for the depolarization process. Activated carbon serves as a catalyst in the synthesis of phosgene from carbon monoxide and chlorine and in the synthesis of sulfuryl chloride from sulfur dioxide and chlorine. The catalytic activity of the carbon is associated with the p electrons in a similar manner as graphite crystals catalyze the trimerization of cyanogen chloride to cyanuric chloride. In each case, weakening of the double bonds in the reactants occurs. In addition, the industrial process in which hydrogen chloride is removed from chloroethane in the presence of  activated carbon presumably has a similar mechanism. By incorporation of heteroatoms the surface of activated carbon can be modified to achieve very high activity for redox reactions such as the decomposition of chloramines and the oxidation of H 2S or SO2. When these catalytic/adsorptive carbons are used for removal of sulfur com-

pounds, the end product is sulfuric acid, which can be washed out with water. Some types of activated carbon are used as catalyst support for industrial syntheses of vinyl acetate and vinyl chloride. Activated carbon is particularly suitable as catalyst support because, in contrast to oxidic catalyst supports, it has no Lewis acidity and therefore does not lead to unwanted polymerization of the vinyl monomers. Vinyl acetate is produced by vinylation of  acetic acid with acetylene in a heterogeneous catalytic gas-phase reaction in presence of a zinc acetate/activated carbon catalyst at 160–240  C. Vinyl chloride can be produced by direct catalytic addition of hydrogen chloride to ethyne at 120–150  C in the presence of an activated carbon impregnated with about 10% of mercury(II) chloride is used as catalyst. Catalysts for various purposes are obtained by depositing noble metals on activated carbon. A well-known example is palladium-coated activated carbon powder [7440-05-3], a typical hydrogenation catalyst for use in suspension. In the Merox process for oxidation of  mercaptans in mineral oil, activated carbons impregnated with cobalt phthalocyanine [3317-67-7] are used.

5. Regeneration and Reactivation The use of activated carbon for gas and water cleaning in industries and environmental protection is particularly economical if suitable processes for the regeneration of spent activated carbon are available. Generally for the regeneration of the spent carbon pressure swing, thermal desorption, extraction, or thermal reactivation processes are used (Table 10) [2–5, 44]. If the adsorbate is bound reversibly onto the surface of the adsorbent the activation energy E des is required for desorption only. For low values of  E des (10–30 kJ/mol), as in gas separation processes, desorption can be achieved by lowering the pressure in a PSA process. For adsorptive removal and recovery of solvent, a higher energy of desorption (30–60 kJ/  mol) is needed for desorption. Usually, superheated steam or hot inert gas is passed through the carbon bed, and this also removes the desorbed substances from the intergranular spaces.

Carbon, 5. Activated Carbon

25

Table 10. Regeneration of spent activated carbon

Process

Basic concept

Main applications

Pressure-swing adsorption Thermal desorption

adsorption at higher pressure, desorption at low pressure steam desorption at 100–150  C, inert gas desorption at 150–250  C elution of organics by solvents or alkali selective gasification of carbonized residues and carbon deposits by steam at 800–900  C

hydrogen purification, air separation (oxygen/nitrogen), methane enrichment from biogas, gasoline vapor recovery solvent recovery, (toluene, alcohols, hydrocarbons, etc.)

Extraction Thermal reactivation

Apart from thermal desorption, extraction with solvents can be used to remove adsorbed substances from used carbons. For example, the elemental sulfur produced in the Sulfosorbon process can be extracted by carbon disulfide. In the literature solvents such as dimethylformamide, acids, and alkali solutions are described [2, 3]. The success of activated carbon in drinkingwater and wastewater treatment has been due to the development of regenerable activated carbon types and the development of effective processes for the regeneration of the spent carbon. Spent activated carbon used in water treatment usually contains many different kinds of  organic substances. A complete thermal desorption of all adsorbed substances is not possible. The decomposition temperatures of high molecular mass compounds are below the desorption temperature. Thus a residual load consisting of pyrolysis products and carbon deposits remains on the internal surface of the activated carbon. These spent activated carbons are thermally regenerated at 800–850  C in a process that is similar to the production of activated carbon and hence more properly called reactivation. Independent of the reactor type used, the spent activated carbon passes through four main steps during the temperature rise to 850  C:  100  C 100–350  C 350–800  C 800–850  C

water removal desorption of low-boiling organic compounds thermal cracking of high molecular mass compounds gasification of the pyrolysis product and carbon deposits by H 2O and CO2

Investigations have shown that carbonaceous deposits formed by thermal desorption treatment of the residual load show a higher reactivity due to their disturbed structure than the carbon surface of the activated carbon. Under

phenols by alkaline extraction, sulfur extraction with CS 2 spent activated carbon from water treatment and other applications

such conditions, the loaded activated carbon can be selectively treated with the water gas shift reaction at ca. 800  C, which restores the original surface structure and adsorption performance of the activated carbon. Drying, desorption, thermal cracking, and gasification can be performed in a single industrial reactor. All types of furnaces used for activation of activated carbon can be used for reactivation. Fluidizedbed reactors, rotary kilns, and multiple-hearth reactors are widely used. Some methods for regenerating powdered carbons have been proposed in the past, but these have not yet attained widespread practical application. To date, powdered activated carbon has only been reactivated in a few cases.

6. Economic Aspects World consumption of activated carbon was estimated at 450 000 t in 1998, and capacity utilization of the activation plants at 70–80%. Percentage consumption (1998) by region was estimated as United States 37%, Europe 32%, Japan 18%, rest of the world 13%. World demand was forecast to rise by 2–3% per annum. Areas of potential growth are flue gas purification in Europe and Japan, water treatment in the United States, Europe, and Japan, and automotive emission control. The regeneration of spent granular activated carbon (Europe: 50 000–60 000 t/a) is a growing market in all areas, but increased regeneration will lead in a fall in the rate of growth in demand for virgin material. Worldwide there are more than 100 activated carbon producers, but the ten largest producers account for about 70% of world capacity. About 45% of production is powdered activated carbon and about 55% is granular and extruded

26

Carbon, 5. Activated Carbon

activated carbon (pellets). The largest producers, divided by country, are given in the following: America USA

Brasil Mexico Europe Belgium France

Germany Netherlands UK Asia China

India Indonesia Japan

Malaysia Philippines

Sri Lanka

Taiwan

Anticarb Barnebey and Sutcliffe Corp. Calgon Carbon Corp. Norit America Inc. Westvaco Corp. Industrias Quimicas Carbomafra Clarimex SA de CV Nobrac Mexicana SA de CV Chemviron Carbon Ceca Pica A.U.G. CarboTech Aktivkohlen GmbH Norit Norit UK   Datong Yuanghua Activated Carbon Plant Ningxia Huahui Activated Carbon Co. Ningxia Longde Activated Carbon plant Shanxi Detong Minerals Bureau Huai Yu Shan Activated Carbon Group Tianjin Anfull Chemical Co. Ltd. Zhejiang Shuichang Activated Carbon Co. Indian Dyestuff Industries Indo German Carbon Ltd. PT. Ikaindo Futamura Chemical Industries Kuraray Chemicals Mitsubishi Chemicals Sankyo Sangyo Takeda Chemical Industries Century Chemical Works Pasific Activated Carbon Cenapro Chemical Davao Central Chemical Philippine Activated Carbon Phileppines Japan Activated Carbon Bieco-Link Carbon Haycarb Tajit China Activated Carbon Industries Taiwan Active Carbon Industries Carbokarn Thailand

References Specific References 1 European Council of Chemical Manufacturers’ Federations (CEFIC): Test methods for activated carbon , CEFIC Brussels 1986, p. 7. 2 H. von Kienle, E. B € ader: Aktivkohle und ihre industrielle  Anwendung, Enke-Verlag, Stuttgart 1980. 3 A. Yehaskel: Activated Carbon , Manufacture and Regeneration, Noyes DATA Corp., Park Ridge, NJ 1978. 4 M. Smisek, S. Cerny: Activated Carbon, Elsevier Publ., Amsterdam–London–New York 1970.

5 P. N. Cheremisinoff, F. Ellerbusch (eds.): Carbon Adsorption  Handbook , Ann Arbor Science. Publishers, MI 1978. 6 R. Ch. Bansal, J.-B. Donnet, F. Stoeckli: Activated Carbon , Marcel Dekker Inc., New York 1988. 7 DE 136 792, 1901 (R. v. Ostrejko). ¨ sterr. Verband f € 8 O ur chem. und metall. Produkte, AT 68 169, 1913. 9 VDI 3674, Waste gas cleaning by adsorption. Process and waste gas cleaning, 1998. 10 V.A. Garten, D.E. Weiss, J.B. Willis, Aust. J. Chem. 10 (1957) 295. 11 H.P. Boehm, E. Diehl, W. Heck, R. Sappock, Angew. Chem. 76 (1964) 742. 12 H. J€ untgen: “Grundlagen der Adsorption,” Staub Reinhalt. Luft  36 (1976) no. 7, 281, 324. 13 H. Seewald: “Technisch verf € ugbare Adsorbentien”, Vortrags€ffentlichung Haus der Technik , vol. 404, Vulkan-Verlag, ver o Essen 1977, pp. 24, 34. 14 W. Kast: Adsorption aus der Gasphase, VCH Verlagsgesellschaft, Weinheim, Germany 1988. 15 S. Brunauer, P.H. Emmet, E. Teller, J. Am. Chem. Soc. 60 (1938) 309. 16 International Union of Pure and Applied Chemistry (IUPAC):  Manual of Definition , Terminology and Symbols in Colloid and Surface Chemistry, IUPAC 1972. 17 R.W. Cranston, F.A. Inkley, Adv. Catal. 9 (1957) 14. 18 H. J€ untgen, M.J. Schwuger, Chem. Ing. Tech. 38 (1966) 1271. 19 ASTM D 3467–99, StandardTestMethod for Carbon Tetrachloride Activity of Activated Carbon, 1999. 20 J. Meunier in: Vergasung fester Brennstoffe und oxydative Umwandlung von Kohlenstoffen, Verlag Chemie, Weinheim, Germany 1962, p. 41. 21 O. Großkinsky, Glu€ckauf  86 (1950) 988. 22 Calgon Corp., BE 839918, 1976. 23 The Carborundum Co., DE 2 624 530, 1976; DE 2 625 625, 1976. 24 H.J. Fornwalt, W.A. Helbig, G.H. Scheffler, Br. Chem. Eng. 8 (1963) 546–550. 25 Kureha Kakagu, Kogyo, DE 2 146 100, 1971. 26 West Virginia Pulp & Paper Comp., DE 1 467 195, 1964. 27 US 2 966 447, 1957 (H.J. Walter). 28 Takeda Chemical Ind., DE 2 400 991, 1982. 29 M.M. Dubinin, W.A. Gordejwa, L.T. Jefimova, B.A. Onusaitis, Z.A. Rjabikowa, Zh. Pril. Kmm. (Leningrad) 43 (1970) no. 10, 2219–2225. 30 C. Hinskelwood, BCURA Gaz. 27 (1956) 3–10. 31 H.J€ untgen, K. Knoblauch, K. Harder, Fuel 60 (1981) Sept., 817– 822. 32 K.-D. Henning, S. Sch€ afer, Gas Sep. Purification 7 (1993) no. 4, 235–240. 33 H. Menig: Luftreinhaltung durch Adsorption , Absorption und Oxidation, Deutscher Fachschriften Verlag, Wiesbaden 1977. 34 J.W. Hassler: Purification with Activated Carbon , Chemical Publishing Co. Inc., New York NY 1974. 35 H. Krill, H. Menig: “L€ osungskonzepte f € ur L€ osungsmittelVermeidung und Verwertung,” Entsorgungspraxis Spezial (1989) no. 6, 20, 26. 36 K. Storp: “Abluft- und Abgasentschwefelung durch Adsorption und Katalyse an Aktivkohlen,” Dechema Monogr. 64 (1970) 91, 94. 37 H.W. Br€ auer: “Abluftreinigung beim Flugzeugbau, L € osemittelr € uckgewinnung in der Wirbelschicht,” UTA Umwelt  Technologie Aktuell (1993) no. 2, 115, 120. 38 K. Storp, Chemiefasern 18 (1968) 173.

Carbon, 5. Activated Carbon 39 E. Richter, E. Pilarczyk, K.-B. Harder, K. Knoblauch, Erd o€l Erdgas Kohle 102 (1986) no. 11, 512–516. 40 W.N. Tuttle, Port Technology International 1 (1995) no. 1, 143, 146. 41 K. Knoblauch, Erzmetall, 33 (1980) no. 2, 109, 114. 42 J. Gottschalk: “Flugstromverfahren–ein Verfahren der Kohlenstoffadsorptionstechnik zur Minimierung der Restemissionen,”  Abfallwirtschaftsjournal 4 (1992) no. 12, 997, 1001.

27

43 K.-D. Henning, K. Keldenich, K. Knoblauch, J. Degel, Gas Sep. Purif. (1988) March, 20, 22. 44 H. J€ untgen, J. Klein, Energy Sources 2 (1976) no. 4, 311–330. 45 K.-D. Henning, J. Degel, J. Klein, K. Knoblauch, gwf-wasser/  abwasser 127 (1986) no. 6, 272–282. 46 R.G. Rice, C.M. Robson: Biological Activated Carbon , Ann Arbor Science Publishers, Ann Arbor, MI 1982.