WO2002038253A2

WO2002038253A2 - Gluing agent for a catalyst

Info

Publication number: WO2002038253A2
Application number: PCT/EP2001/013002
Authority: WO
Inventors: Christophe Lahousse; Caroline Cellier; Paul Grange
Original assignee: Universite Catholique De Louvain
Priority date: 2000-11-09
Filing date: 2001-11-09
Publication date: 2002-05-16
Also published as: WO2002038253A3; AU2002221835A1

Abstract

According to a first aspect, the present invention relates to a composition consisting of an active catalytic phase and a support, wherein said active phase is deposited onto said support by means of a gluing agent. The invention comprises organic and inorganic binders as said gluing agents. In particular, the invention relates to the use of polyfurfurylalcohol (PFA) as organic binder, preferably PFA which undergoes a two-step calcination process to obtain full activity. The invention furthermore provides a mixture of sepiolite and ludoxÒ as organic binder to glue a catalytic material to a support. The present invention also refers to MnO2, preferably g-MnO2 as said catalytic material. According to a second aspect, the present invention also relates to the development of a new industrial shape for monoliths. According to a third aspect, the present invention describes the use of g-MnO2 as very selective catalyst for removal of Volatile Organic Compounds (VOC's) containing heteroatoms like nitrogen, sulphur or halogens, and in particular of nitrogen-containing VOC's.

Description

GLUING AGENT FOR A CATALYST

This invention relates to the use of a gluing agent to bound an active phase, for example catalytic material, on conventional and unconventional supports. The present invention also relates to the development of a new industrial shape for monoliths. Still another aspect of the present invention is to provide a selective catalyst for removal of Volatile Organic Compounds containing heteroatoms like nitrogen, sulphur or halogens, and in particular nitrogen-containing VOC's.

It is known that heterogeneous catalytic oxidation is an industrially useful process. Both a very performant active phase (catalytic material) and an adapted support are necessary in order to make an efficient industrial heterogeneous catalyst. The active phase provides the activity, selectivity and very often determines the lifetime of the system, whereas the support provides mechanical resistance and the macroscopic shape. Of course, when the support is not constituted by the active phase, the active phase needs to be bonded onto the support in such a way that it preserves its catalytic capacities. In addition, the shape of the support should be opportune in order to optimize contact between the fluid and the solid catalyst. Monolith structure constitutes a typical example and is explained further below.

As an active phase generally a noble metal (Ru, Pd, Pt, Au...) is used, or a high surface area cobalt oxide. In addition, MnO₂, CuOx and CoOx can be used as well.

The supports which are generally used are pellets or monoliths. Pellets are generally constituted of alumina supports. The pellets can receive an active phase by precursor impregnation and subsequent transformation. A monolith is a rigid ceramic or metallic structure which is formed by the assembly of straight smooth thin wall channels. By the assembly of these channels, the macroscopic surface of contact between the solid catalyst and the reactant fluid is maximized, and the resistance to flow (pressure drop) is minimized. Monoliths are notably used for removal of Volatile Organic Compounds (VOC's). Volatile

Organic Compounds are major atmospheric pollutants through their contribution to ozone formation. Moreover, these compounds are responsible of almost all odor nuisances generated by industrial sites, waste and farming activities.

In general, ceramic monoliths are prepared by extrusion using AI(OH)₃ as a binder and HNO₃ as a peptizer. This type of monoliths can be reinforced with an organic binder filling cracks, or alternatively a clay like sepiolite can be used as binder. Deposition of the active phase onto the whole monolith is comparable to the deposition in pellets. However in case of monoliths, it is important to obtain a uniform dispersion throughout the monolith structure.

Metallic monoliths are obtained by waffering a metal sheet and rolling it onto a flat one. To obtain sufficient binding of the active phase, oxide anchoring layer are generally grown out of the metal surface. In the case of steel (FeCrAI alloy), the layer is obtained by treating the structure at high temperature during several hours. Another method for the development of an aluminium monolith is by electrolytic oxidation of aluminium.

Others forms of structured supports which were developed are ceramic foams, metal grids, metal grid stackings and carbon tissues. In US Patent 4,157,315 a method was disclosed for bonding active catalytic material to a support wherein the bonding is carried out in aqueous medium in the presence of at least two water-soluble flocculating agents. One of these agents has a preferable affinity for the active phase, while the other has preferably affinity for the support. When deposited upon the support, they form a water-insoluble adhesive comprising bonding the active phase to the support. However, in this method one of the flocculating agents react with, or have affinity for the active substance.

The problem underlying the present invention is that in the preparation of an industrial heterogeneous catalyst, it is difficult to form satisfactory and adherent coating of the active phase on the support.

In the conventional preparation of a catalyst, adhesion is only provided by the interaction between the active phase or catalytic material with the support. In order to obtain sufficient binding of the active phase to the support, the adhesion procedure needs to be optimized for each support material and each active phase, which is on itself time and money consuming. Moreover, by this method the mechanical resistance of the deposit will be that of the active phase. Despite all efforts, some interesting active phase could remain unusable if no method can be found to bond this in adequate amount and to keep its activity on the wanted support.

Therefore, the first main object of the present invention is to provide a method for bonding an active phase onto a support in a relatively inexpensive manner.

The second main object of the present invention was to provide a new industrial shape for monoliths, suitable as a preferred support. In the last 20 years, application of monolithic carriers enabled fast development of catalytic methods for reduction of atmospheric pollutants generated by combustion of fuels in the power industry, in the heat generation industry, in internal combustion engines, in processes of utilization of wastes, etc.

Monoliths are generally constituted by an assembly of straight channels. In the case of metallic monoliths, such straight channels are obtained by waffering a plate of metal and rolling it onto a flat metal plate.

US Patent 5,681 ,538 is related to a monolith assembled from a plurality of plates that are mounted one atop the next. The plates are generally rectangular and define a central baffle portion having a pair of substantially straight, parallel sides and having attached to the baffle portion at the ends thereof a support flange which is bendable upward into a configuration that enables it to support a second plate in spaced, parallel relation to the first plate, and anchor flanges that are bendable downward, to seal the corners on the monolith and, optionally, to engage the support flanges of one or more underlying plates.

All these monoliths, whether assembled or not, consist of straight channels, and are further referred to as unidirectional or smooth monoliths. These straight channels allow to minimize 5 pressure drops but favour the onset of a laminar flow.

However, the laminar flow present in such classical monoliths limits diffusion of the reactant towards the catalyst surface (due to the thermal agitation), resulting in lower catalytic performances.

Therefore, the present invention provides a new industrial shape for monoliths whereby the o aforementioned disadvantage of unidirectional monoliths can be obviated.

Monolith structures are notably used for VOC removal. Volatile organic compounds (VOC's) are pollutants because, in addition of being odorous or toxic sometimes, they almost always contribute to ozone formation (Horsley JA, Catalytica Environmental Report E4, Catalytica Studies

Division, Mountain View, CA, 1993). Moreover, these compounds are responsible of almost all 5 odour nuisances generated by industrial sites, waste and farming industries (Chem. Ind. 23: 855

, 1991; Milmo S, Chem. Mark. Rep. 238: 18, SR16, 1990). Accordingly, when VOC emissions cannot be avoided, they ought to be controlled.

In order to remove these undesirable substances, several technologies have been developed. Among them, catalytic combustion is very attractive due to its advantage of higher o selectivity (towards complete oxidation) and tower reaction temperature than thermal incineration

(Spivey JJ, Ind. Eng. Chem. Res. 26: 2165, 1987). Both noble metals and oxide catalysts have been studied to achieve this goal.

Lahousse et al. (Journal of catalysis 178:214, 1998) have shown that γ-MnO₂ is a very promising catalyst to treat VOC emitted by a printing plant. In this study, Lahousse et al. did 5 compare the performances of two very active catalysts (one metal oxide γ-MnO₂ and one noble metal catalyst Pt/TiO₂, which consists of platinum supported on titaniumdioxide (TiO₂)) for VOC removal (ethylacetate, hexane and benzene). This comparison took into account not only the activity but also the sensitivity to competition effects between compounds, the influence of water vapour and the stability. This investigation showed that the metal oxide catalyst γ-MnO₂ proves to 0 be more active than the supported noble metal one. The complete combustion of various VOC's is obtained at much lower temperature than with commercial catalysts. Moreover, its performance is less affected by interferences between VOC's than those of the noble metal catalyst.

On the other hand, nitrogen-containing VOC's have received less attention, while these compounds are present in farm air and generally produce an offensive odour. The catalytic 5 oxidation of nitrogen-containing VOC's (N-VOC's) has the additional problem that destruction of these compounds results in nitric oxide (NOx) formation. NOx are strong pollutants and contribute 10 times more than VOC to ozone formation. NOx formation must therefore be limited as much as possible.

Moreover, most of the commercial available combustion catalysts which are used for N- VOC still produce too high NOx concentrations. Therefore, the present invention provides a γ-MnO₂ catalyst which is more appropriate for the treatment of N-VOC's, and also S- and halogen- containing VOC's.

In summary, the first main object of the present invention is to provide a method for bonding an active phase onto a support in a relatively inexpensive manner. It is another object of the invention to provide a method which allows bonding of the active phase onto the support in a straight-forward way. Yet, another object of the invention is to provide a method which allows to bind any type of active phase onto any type of support. In addition, another object of the invention is to provide a method for binding of the active phase onto the support, whereby the catalytic activities of the active phase are at least equally preserved or preferably even more active after deposition.

The second main object of the present invention is to provide a new industrial shape for monoliths.

The third main object of the present invention is to provide a selective catalyst for removal of VOC's containing heteroatoms like N, S or halogens, and in particular N-VOC. All the objects of the present invention have been met by the embodiments as set out below.

One aspect of the present invention provides gluing agents which enable to bind an active catalytic material to any type of support. These gluing agents allow to deposit any suitable amount of active phase. In addition, by using a gluing agent, unconventional, very moldable or cheap unprepared supports can be used, with the result that the mechanical and catalytic properties of the deposit obtained are equally or superior to that of the active phase. Thus, the use of such gluing agents provides intrinsic advantages. In conventional deposition, the mechanical properties of the deposit are those of the active material. Surprisingly, when using a gluing agent according to the invention the catalytic properties of the active phase are reinforced by said binder. As a result, incompressible dusty catalysts like MnO₂ can be used. In addition, synthesis of the catalyst must not be performed in situ. The synthesis of the catalyst can even be optimized without taking into account support accessibility. With a gluing agent, the deposition of the active phase can be performed prior to monolith formation. The active phase can be homogeneously distributed all over the catalyst. It is difficult to obtain such an even distribution by conventional slurry impregnation methods. When using a gluing agent, catalyst deposition can be performed using conventional "painting" method, namely mature processes easily adaptable to any case. According to the invention, both organic as inorganic binders can be used as said gluing agents.

Organic glues may be used to bind the catalyst on any support provided these glues actually possess gluing properties and retain these properties up to the range of temperature in which the catalyst is active. In addition, the organic glue should not block any access of the reactant to the surface, or otherwise it should be able to treat it in such a way that access to the catalyst is obtained. Rather viscous glues that stick the catalyst particle without diffusing into the pores can provide this result. According to the invention, polyfurfuryl alcohol, polyacrylonitrile, polyvinylidene chloride, epoxy glue, araldite^® and/or mixtures thereof are good examples of such organic glues. The use of organic glues provides an alternate choice to inorganic ones and present several specific advantages. Compared to inorganic binders, polymerized glues will be (like most polymers) very resistant to acid or basic attack in aqueous medium. In addition, a non-calcined organic bounded deposit is extremely adherent and moldable. This allows to do the waffering after the deposition in case of a monolith. Thus the active material can be easily homogeneously distributed on the flat surface, whereas in the case of deposition on an already waffered plate, one always has to avoid accumulation in holes and poor deposition in bumps leading to partially or totally plug monolith. In general, due to its excellent adherence and deformabiiity, an organic binder is preferred each time imprinting or important deformation of the covered support is required. Finally, using an organic binder allows to obtain adherence on "dirty" surfaces. On aluminium for instance, the poorly adherent alumina layer present on any aluminium surface results in weak adhesion of inorganic bounded deposit. When using an organic binder as heated polyfurfurylalcohol (PFA), the viscosity of the binder is not too important since the binder will diffuse around the "dirty" particles down to the bulk support. As a result, excellent adhesion can be created on non-treated aluminium.

More particular, one specific embodiment of the present invention is based on the use of polyfurfurylalcohol (PFA) as organic glue. PFA is a polymer that is used to fill cracks in monoliths and to bind carbon particles together. For these purposes, this polymer is diluted in acetone and then carbonized at high temperature (T>600 °C). As mentioned above, PFA has the advantage being a viscous organic material that adheres on any clean or dirty surface. In addition, since it is rather viscous it will stick the catalyst particle without diffusing into the pores, and therefore will not block any access of the reactant to the surface.

According to another feature of the invention, inorganic binders can be used as said gluing agents. Versatile inorganic binders are binders relying on unspecific links like van der Waals and capillary forces, H-bonding, and mechanical anchoring in supporting bumps and holes. In order to increase the versatility of the binder, different components creating up to different levels and different bonding procedures can be used in combination. The binding composition should be able to glue the catalyst particles on the support and together. Binders that can be applied in liquid to slightly viscous forms and that turn into solid upon treatment will favour mechanical anchoring. In all

5 cases, the binders should not block any access to the catalyst.

Preferred examples of such binder components are colloids, clays, concrete and cement components, sodium silicate and peptized alumina.

Colloids will provide adhesion through capillary forces resulting in a rigid deposit. Preferred typical colloids are silica colloids like ludox^®, alumina colloid (Alfa12733) and similar colloids.

[0 Clays offer many OH-bonding possibilities and binding to fiber like particles enabling to bind a thick layer of catalyst particles. Sepiolite, bentonite, montmorillonite, ataplugite and vermiculite are examples of preferred clay materials; related material like LDH or pillared clays offers the same characteristics.

Concrete and cement components can provide capillary anchoring through a complex

[5 solubilization and crystallization process (e.g. calcium silicate). Sodium silicate is a well-known inorganic glue with strong adhesive power. Peptized alumina is widely used as a binder with some acid peptizer for bulk monolith. Alumina can be used to prepare thick washcoats.

One specific preferred embodiment of the present invention is based on the use of a mixture of sepiolite and ludox^® as said inorganic glue. Sepiolite is a clay which is used to produce ceramic

20 monoliths. Ludox^® is the trademark of a series of silica colloids in aqueous solutions. By mixing these components with water and a catalytic material, it is possible to prepare a slurry that will adhere on any surface and yield an active phase retaining full activity. Such deposition has been successfully performed on steels, activated aluminium, metal grids, carbon tissues, domestic hood filters, heating wire (resistance), glass, Teflon and polyurethane foam. This mixture can also be

25 used to prepare bulk pellets (extrudates) and possibly monoliths. Deposition can also be performed on bare aluminium, however, due to permanent covering of this metal by a poorly adherent oxide layer a weaker adhesion was obtained on this material. Taking into account the immediate and surprising success of all the trials, this method can be recommended to bind a suitable catalyst on any suitable type of support. Non-limiting examples of material suitable as support include,

30 ceramics, organic and inorganic polymers, glasses, metals, carbons, paper, cardboard, polymeric film, glass beads, carbon granules, glass wool, metal sponge, etc. Said support can take the form of flat sheets, tubes, honeycomb structures, meshes of relatively high surface area, sieves, rings, saddles, etc.

The deposit possesses good mechanical resistance. For instance, two metal sheets of (one

35 flat and one waffered) can be rolled together into a monolith without loosing matter. Flexible support such as polyurethane foam looses its flexibility when covered by active phase. In example 1 , the apparent activity of bulk γ-MnO₂ pellets and γ-MnO₂ bounded onto pellets by using an inorganic glue such as a mixture of sepiolite and ludox^® is compared. From this example it is clear that the deposit which is obtained when using sepiolite+ludox^® as gluing agent is more active (due to less diffusional limitation) than the pure catalyst.

5 In addition, using sepiolite and ludox^® as gluing agents has several advantages. First, a mixture of sepiolite+ludox^® is a very versatile binder and therefore binding can be obtained onto any support, presumably with any catalyst, in any quantities. Highly diluted slurry can be used to form a thin film, superposed deposit or bulk extrusion can be used to form mass deposition. Second, any conventional or unconventional covering method can be used (slurry impregnation, batch o impregnation, paint brush deposition etc.). Good extrusion can be obtained at very low compression. Furthermore, with this binder it will be possible to insert a catalyst in any existing structure (e.g. to paint the tubing of an evacuation pipe). Controlled drying can improve the mechanical properties but no high temperature thermal treatment is required. Thus, new support material such as polyurethane foam can be obtained. In addition, as a water paint, sepiolite+ludox^®

5 slurries do not contribute to production of Volatile Organic Compounds (VOC's), and in addition can be easily cleaned.

Examples of suitable catalysts according to the invention include but are not limited to metal oxides selected from titanium, chromium, iron, cobalt, nickel, copper, zinc, molybdenum, tungsten, tin and the like or mixtures thereof. Furthermore the catalyst may comprise one or more noble

-0 metals (Ru, Pd. Pt. Au...).

In a more specific embodiment the present invention relates to the use of MnO₂, preferably γ-MnO₂, as said active catalytic phase of the invention as defined above. The inventors previously demonstrated that MnO₂ deposited without any gluing agent, onto e.g. pellets, is far more active than commercial catalysts (Lahousse C, Journal of Catalysis 178:214, 1998). The complete

!5 combustion of various VOC's was obtained at much lower temperature than with commercial catalysts. They furthermore compared the catalytic activity of various forms of Mn dioxides towards

VOC's oxidation, and found that the γ-form of MnO₂ is of superior activity (Lahousse C. in 3^rd World

Congress on Oxidation Catalysis, Stud. Surf. Sci. and Catal., 110:777, Elsevier, Amsterdam, 1997).

The same results are now confirmed with gluing agents according to the invention used to

50 bind γ-MnO₂ onto a support. By using said gluing agents the macroporosity of the deposited γ-MnO₂ is increased, and a higher internal diffusion capacity is obtained. As a result γ-MnO₂ deposited by means of a gluing agent is far more active when compared to conventional deposition without any gluing agent. This is further illustrated in example 1 which compares the activity of bulk γ-MnO₂ pellets and γ-MnO₂ bounded onto pellets by using an inorganic glue such as a mixture of sepiolite

55 and ludox^®. In another specific embodiment of the present invention, the inventors describe the use of PFA to glue the catalyst on a support, provided a two-step calcination procedure is performed. Calcination is required to recover the accessibility of the catalyst and as a consequence its activity. Thus, it is possible to obtain with this method a catalyst which strongly binds to the support and retains full activity. The use of PFA as gluing agent is further illustrated in example 2, where the activity of γ-MnO₂+PFA deposited on aluminium strips is compared to the activity of the same weight of γ-MnO₂ particles alone.

Another advantage of using PFA is that several layers of pure PFA and catalyst can be superposed with full preservation of the activity of all layers. Moreover, PFA creates a highly flexible deposit. The covered metal surface can be waffered to form a monolith or can be molded to any desired shape. Several layers can be superposed, any quantity of catalyst can be deposited, and in addition, the activity is solely limited by internal and external diffusional limitations. The system has been tested in gas phase applications but could also applied to aqueous medium.

In example 4 the inventors demonstrate that, by using either PFA or a mixture of sepiolite+ludox^® as gluing agents, they have been able to produce usable monolith structures of different forms, and to implement them at pilot scale.

The second main object of the present invention was to provide a new industrial shape for monoliths.

In this aspect, the invention provides a new monolith shape consisting of a bilayer of which one layer has an unidirectional structure, whereas the second layer consists of a relief structure.

In one specific embodiment of the present invention, the relief structure of the second layer consists of a bidirectional pattern.

In yet another embodiment of the present invention, this bidirectional pattern is made by folding the second layer in 2 directions, perpendicular onto each other. Thus, simply by folding the original metal plates before waffering and rolling it is possible to make folded channels. Folded monoliths allow to optimize the contact between the gas and the catalyst. With this structure, very low pressure drops can be maintained at high space velocities like in straight unidirectional monoliths. But in addition a flow turbulence (similar to what is obtained with grid stacking) is created, which results in a dramatic increase of performances at high space velocities.

This is further illustrated in example 3 where the conversion of hexane by γ-MnO₂ is compared, when this is deposited onto either a unidirectional smooth or a bidirectional, folded monolith. A dramatical increase of activity is recorded when using such folded monolith structure. This increase of activity can be attributed to the transformation of the laminar into turbulent flow. With a turbulent flow the transfer of the reactant towards the gas phase onto the catalyst washcoat is provided not only by the thermal agitation (like with the laminar flow) but also by the gas turbulence. Appearance of external diffusional limitation can be postponed and the activity can be maintained at higher space velocities. This improved activity is obtained at the expense of some increased pressure drop.

In summary, monoliths with such folded channels preserve a turbulent flow which results in a dramatic increase of performances at high space velocities.

Finally, the third main object of the present invention a selective catalyst for removal of VOC's containing heteroatoms like N, S or halogens, and in particular N-VOC's. In that aspect, the present invention relates to the use of γ-MnO₂ as a very effective catalyst for the removal of N -containing VOC's. In a more specific embodiment, the invention relates to the use of γ-MnO₂ as a catalyst for the combined removal of N-, S- and halogen-containing VOC's.

The N-VOC present in livestock are mainly primary, secondary and tertiary amines (O'Neill DM, Phillips VR; J. Agric. Eng. Res. 53: 23, 1992). Several studies on the complete oxidation of amines have shown that both noble metal and metal oxide catalysts can be considered for this purpose. In this aspect, the inventors have compared NOx production during the combustion of monomethylamine for most of the possible combustion catalysts (noble metal (Pt, Pd, Ag and Ru) supported on titanium dioxide and pure or doped oxide catalysts (γ-MnO₂, CoO_x, NiO_x, FeO_x, CuO_x)). This study, as further explained in example 5, showed that γ-MnO₂ is the most active and in addition is the catalyst which produces the less NOx.

The followed examples are given for purpose of providing those skilled in the art with a better understanding of the invention. When MnO₂ is cited in these examples, the inventors used the γ-form of MnO₂. In the legends of the figures, MnO₂ equally refers to γ-MnO₂.

EXAMPLE 1: Use of inorganic binders, in particular a mixture of sepiolite and ludox^® to glue γ-MnO_z as catalytic material to a support.

Procedure for gluing catalyst to a support by means of sepiolite and ludox^®- In this method a slurry of tuneable viscosity is prepared and spread over the desired support. For monolith preparation, the following procedure is used as outlined hereafter. 300 g of distilled water is weighed and mechanically agitated. Vigorous mechanical agitation is maintained throughout the preparation and spreading time. Then, 1 g of NaOH 1 M or 1 g of cone. NH₄OH is added to the solution, followed by the addition of 50 g of ludox^® SM 30 (Aldrich 40.079-4). Immediately thereafter 62.5 g of sepiolite (EXAL given by TOLSA benelux) are mixed with the solution, followed by 500g of γ-MnO₂. During the addition of γ-MnO₂ the stirring rod is moved throughout the bath to ensure that all the catalyst is well suspended in the slurry.

The obtained paste can then be painted over flat and waffered plates. If painting exceeds more than about 4 hours, 10 g of water should be added in order to moisten the "dried" paste. Covered plates are then rolled one onto the other to form a monolith.

Full water elimination by controlled drying will increase the mechanical resistance of the deposit. Drying must however be slowly performed in order to avoid sudden and excessive water vapour formation leading to sintering. The temperature should be raised by about 0.25 to 0.75 °C per minute and maintained at said final temperature during about 10 to 50 min until a temperature of about 100 to 150 °C is reached. Preferentially the temperature is raised by 0.5 °C per minute until a temperature of 120 °C is reached, and maintained at said final temperature during 30 min

When a less viscous slurry is appropriate, for instance for thin film preparation and for impregnating flexible supports (e.g. polyurethane foam, hood filter, carbon tissue etc.) 700 g of water can be used instead of 300 g. Conversely, a thicker paste can be prepared with only 219 g of water, for preparing extrudates or for bulk monolith extrusion.

Figure 1 compares, in the absence of external diffusional limitations, the apparent activity of bulk γ- MnO₂ pellets and γ-MnO₂ bounded onto pellets by using an inorganic glue such as a mixture of sepiolite and ludox^®. In this example, the activity of γ-MnO₂ as catalytic material refers to its conversion efficiency towards hexane. The activity of bounded γ-MnO₂ pellets having up to 0,8 mm diameter is conserved, while the apparent activity of bulk γ-MnO₂ pellets of the same size is dramatically reduced. EXAMPLE 2: Use of an organic binder, in particular PFA to glue γ-MnO_z as catalytic material to a support.

Synthesis of PFA (C₅H₆θ₂). The synthesis is done under magnetic stirring on a hot plate. 685 ml of distilled water is heated to temperature (T) >95°C. 600 ml of furfuryl alcohol (Aldrich 18,593-0) is added. The mixture is stirred, heated until it reaches a T>95°C and 1 ,5 ml of concentrated sulfuric acid (Merck 96% suprapur^®) is added. One minute after addition an "explosive" polymerization reaction occurs. This mixture is kept under the same heating and stirring conditions for exactly 10 min and is then removed from the hot plate. Once the temperature has decreased below 60°C (at least 2h), the aqueous phase is roughly separated (poured slowly out of the beaker) and the organic viscous PFA is filtered on a 1 mm opening metal grid. The remaining drops of aqueous phase floating on the PFA surface are carefully removed (1 ° poured, 2° picked using a pasteur's pipette, absorbed on cleaning paper (Tork)). The PFA can be stored in a bottle until use.

Procedure for gluing catalyst to a support by means of PFA.

After carefully removing the last drops of aqueous phases, the PFA is painted with a paintbrush onto the selected support (for monolith: a commercial aluminium foil having 0,1 mm thickness). The two faces are glued and the remaining binder is wiped (adsorbed) using a cleaning paper (Tork rolls of paper). The glued support is passed in a bath containing the powder catalyst (γ-MnO₂) and introduced in an oven at 120°C for about 5 minutes. This treatment decreases the viscosity of the binder which surrounds the catalyst particle. The hot covered support is then quickly plunged in the catalyst powder bath again. Up to 70 g of catalyst per m² of support can be deposited at once. To obtain higher loading, the first deposit is glued and plunged again in the catalyst bath, followed by 5 minutes heating and impregnation as described above. Following this procedure, up to 4 layers of binder and catalyst have been accumulated on supports of 15 m length and 0,1 m width. These samples retained between 160 and 250 g of γ-MnO₂ per m². The covered plates can be waffered without loosing any catalyst. By rolling together both flat and waffered plates a cylindrical monolith having 100 mm diameter and height has been prepared.

Procedure for PFA controlled decomposition.

The monoliths that have been prepared need to be calcinated in order to be active. The aim of this calcination procedure is to make the catalyst burn most of the binder in such a way that the gaseous reactant becomes accessible. This calcination should occur via a controlled procedure in order to 1 ) retain enough organic material to hold the catalyst on the support surface and 2) avoid a too violent combustion leading to high temperature generation which would destroy the catalyst. The temperature ramp should be very slow and very well controlled, which can be done by using furnace ventilation of a gas chromatograph. The monolith should be calcinated in an upright position allowing access to the air below (we used ceramic crucibles having approximatively 1 cm height). In addition, the monolith should be protected from the furnace ventilation (we used a beaker returned around the monolith). Two successive calcination steps are required to ensure that all the remaining binder is burned. Variations in binder properties result in different optimal maximum temperature. In case of very viscous PFA, some incomplete decomposed binder will remain which leads to monolith destruction under uncontrolled heating. Therefore, in order to average the binder properties, a two ramps calcination is required. For γ- MnO₂ monoliths, the optimum temperature profile is as followed. In a first calcination step, the PFA is heated (calcinated) by raising the temperature by about 2 to 10°C per minute until it reaches the range of about 130°C to 180°C, followed a slower increase of about 0.05 to 0.5°C per minute until a temperature in the range of about 240 to 300°C is reached, and maintained at said final temperature during about 40 to 80 minutes. The PFA is then allowed to cool down until 120 to 170 X, followed by a second calcination step where temperature is raised again by about 0.05 to 0.5°C per minute until a temperature in the range of about 240 to 300°C is reached, maintained at said final temperature during about 40 to 80 min. Preferentially the temperature is raised by 5°C per minute until it reaches 150°C, followed a slower increase of 0.1 °C per minute until a temperature of 270°C is reached, and maintained at said final temperature during 60 minutes. The PFA is then allowed to cool down until 150°C, followed by a second calcination step where temperature is raised again by 0.1 °C per minute until again a temperature of 270 °C is reached, and maintained at said final temperature during 60 minutes.

Following the procedure as described above, 10x10 cm cylindrical monoliths having up to 500g γ- MnO₂ can be prepared. The active deposit in these monoliths retains a good adherence and mechanical resistance, equally to the specific activity of γ-MnO₂ .

In figure 2 the activity of γ-MnO₂+PFA deposited onto aluminium strips is compared to the activity of the same weight of γ-MnO₂ particles (diameter 0.2-0.315 mm). As shown by this figure, the activity of γ-MnO₂+PFA is fully conserved. EXAMPLE 3 : Comparison of the activity of γ-MnO? as catalytic material supported onto an unidirectional or a folded, bidirectional monolith.

The results given in figure 3 represent the conversion of hexane by γ-MnO₂ when either an unidirectional monolith (silica alumina) or a bidirectional, folded monolith is used as support, γ- MnO₂ is bonded onto the support by means of the mixture of sepiolite and ludox^®. Although the two monoliths in this comparison possess the same amount of catalytic material and the same thickness of deposit, their activities are very different which can undoubtedly be attributed to differences in external diffusional limitations. As shown by the non-exponential behavior of the curve obtained with the non-folded monolith, the apparent activity of this catalyst is dramatically limited by diffusional limitations, whereas the folded monolith is much less affected. As a result, 70% conversion is obtained at 180°C on the folded monolith while this conversion is only obtained at 220°C using the straight monolith. Over 95% conversion is obtained at 220°C with the folded monolith that would probably never been obtained with the straight monolith. At a given space velocity, folding increases the maximum achievable conversion.

EXAMPLE 4: Use of monoliths of different shapes, prepared using sepiolite+ludox^® or PFA as gluing agents, in the conversion of hexane by y-MnO? at pilot scale.

Lahousse et al. recently confirmed the efficacy of γ-MnO₂ as catalyst for VOC removal at pilot scale with monolith structures used as support (figure 4). Four types of monoliths were used, either a commercial Pt monolith, a folded or a smooth γ-MnO₂ + sepiolite+ludox/aluminium monolith and a γ-MnO₂ + PFA/straight monolith. Next the conversion of hexane was compared for all type of catalysts. This comparison showed that at pilot scale at 220°C over

95% conversion of n-hexane can be obtained with a folded γ- MnO₂+sepiolite+ludox/aluminium monolith, about 70 % with a smooth γ-

MnO₂+sepiolite+ludox/aluminium monolith, about 63 % with a γ-MnO₂ + PFA/straight monolith whereas a commercial monolith constituted of noble metal Pt on FeCrAlloy barely achieve this level of conversion at 370°C. At 220°C, the commercial Pt monolith only converts 30% of the reactant.

EXAMPLE 5: Conversion of monomethylamine by both noble metal and oxide catalysts.

Noble metal (Pt, Pd, Ag and Ru) supported on titanium dioxide (TiO₂) and oxide catalysts (γ-Mn0₂,

CoO_x, NiO_x, FeO_x, CuO_x) were prepared. Catalytic tests with 250 ppm of monomethylamine in air were performed in order to compare the oxidation activities of these different catalysts. Activity scales for metal noble, metal oxide catalysts were established, and the destructibility of the monomethylamine was evaluated on both families of catalysts.

Figure 5A represents the conversion of one typical nitrogen-containing VOC, namely monomethylamine, as a function of the reaction temperature on the most commonly used catalysts. As shown by this graph γ-MnO₂ achieves complete conversion at 160 °C whereas the other catalysts only obtain this result at 80°C or higher. It is clear that γ-MnO₂ is the most active catalyst for nitrogen-containing VOC deep oxidation.

Figure 5B (+ inset) represents the total amount of NOx (NO+NO₂) produced during monomethylamine combustion as a function of the reaction temperature. For most of the catalysts, the total amount of NOx increases till 100% conversion is reached, corresponding to an increase of

NO formation. At 100% conversion of the reactant, the NOx amount suddenly decreases and then dramatically increases when the reaction temperature is raised well above the temperature at which

100%) conversion is obtained. The second increase is mainly due to overoxidation of the nitrogen into NO₂. As shown by this figure 5B (+inset), only a few catalysts show both a limited NO formation during conversion and no increase of NOx formation after 100% conversion. Among these catalysts, γ-

MnO2 forms the less NOx.

EXAMPLE 6: Resistance of glued catalyst to attrition in pilot test with industrial feed.

Using example 1 procedure, a monolithic structure consisting of a stacking of a hundred stainless 304 grid was prepared. This monolith was submitted to the emissions of a car painting booth for more than 1000h hours. Despite the presence of dust and pigments in these emissions no decrease of activity was detected during the whole test indicating that no catalyst attrition occurred. This experiment demonstrates that glued catalyst possesses enough mechanical strength to withstand industrial conditions.

EXAMPLE 7: Applicability of the inorganic gluing agent to any water insoluble solid.

In order to demonstrate the applicability of the inorganic gluing agent to bind any solid on any support, SiO₂₎ and AI₂O₃, two widely used catalyst have been glued to a polyurethane foam. Bonding was readily obtained by using the procedure and composition used for MnO₂ and simply correcting the amount of water by adding (or subtracting) the difference of pore volume between MnO₂ and the replacement solid to glue.

For instance, when taking as a basis a slurry constituted of 50g MnO₂ + 26g H₂O + 5g ludox SM30 and 6.25 sepiolite, for deposition on a polyurethane foam, deposition of AI₂O₃ Aldrich 26,774-0 and Si0₂ silica gel Aldrich 24398-1 was readily obtained by simply measuring the amount of water to be added to 50g of these solid to obtain a slurry type mixture.

For example, 17g of water are necessary to obtain a slurry from 50g MnO₂ 39 g of water are necessary to obtain a slurry from 50g AI₂O₃ 173g of water are necessary to obtain a slurry from 50g of Silica gel

26 +(39-17)= 48 g of water resulted in as good deposition of AI₂O₃ as the one obtained with MnO₂ As confirmed by the good deposition obtained with Silica gel using 26+173-17=182g of water, the water amount correction can be actually applicable even to solid of very different porosity.

EXAMPLE 8: Conservation of the catalytic properties of glued MnO? for any reaction.

In example 4, the conservation of γ-MnO₂ activity for VOC destruction was illustrated. The conservation of activity was also confirmed for any other reaction like demonstrated in this example.

A well known catalytic application of MnO₂ is ozone destruction. In this example, the activity in ozone decomposition of a MnO₂ sample deposited on polyurethane foam has been measured.

Ozone was generated by arc discharge in a volume. The concentration of ozone emitted from this volume for a given diluting air flow has been measured. Once the maximum concentration had passed (and the evolution of the concentration could be inferred), the outgoing air flow was passed through a reactor containing either a glass blank catalyst or MnO₂ glued on polyurethane foam. The results obtained are displayed on figure 6, wherein it can be seen that the activity of glued MnO₂ for ozone destruction is clearly conserved. The conservation of the activity was not reaction sensitive

EXAMPLE 9: Catalyst for odor treatment.

As shown in figure 5B, MnO₂ combines efficiency and selectivity in the treatment of odorous VOC such as N-and S-containing VOC. These laboratory results have been confirmed by the measurement of odor abatement performed on an industrial feed. Indeed, during pilot scale treatment of emission produced by an industry working with bitumen, measurements of odor concentration in treated and non treated gas were performed and showed that γ-MnO₂ monolith can actually abate odor efficiently. The odor level was decrease to the level of a blank sample, and a diminution of the odor level of 78% + 22% was measured.

FIGURE LEGENDS

Figure 1

Conversion of hexane as a function of the reaction temperature and particle size, when using bulk 5 γ-MnO₂ pellets and γ-MnO₂ bounded onto pellets by using an inorganic glue such as a mixture of sepiolite and ludox^®.

Figure 2

Conversion of hexane as a function of the reaction temperature, when using bulk γ-MnO₂ and γ- o MnO₂ +PFA deposited onto aluminium strips.

Figure 3

Conversion of hexane as a function of the reaction temperature, when using γ-MnO₂ + sepiolite+ludox deposited onto either a bidirectional folded or a unidirectional, straight monolith. 5

Fi ure 4

Conversion of hexane as a function of the reaction temperature, when using either a commercial Pt monolith, a folded or a smooth γ-MnO₂ + sepiolite+ludox/aluminium monolith and a γ-MnO₂ + PFA/straight monolith.

!0

Figure 5

(A) Conversion of monomethylamine as a function of the reaction temperature, by both noble metal and oxide catalysts.

(B) (+inset) Total amount of NOx (NO+NO₂) produced during monomethylamine combustion as »5 function of the reaction temperature, when using both noble metal and oxide catalysts.

Figure 6

Effect of the MnO₂ catalyst on ozone concentration.

Claims

1. A composition consisting of an active catalytic phase and a support, wherein said active catalytic phase is deposited onto said support by means of a gluing agent.

2. A composition of claim 1 wherein said gluing agent is an organic binder or inorganic binder.

3. A composition of claim 1 or 2 wherein said gluing agent is an organic binder selected from the group consisting of furfurylalcohol, epoxy glue and araldite^® and/or polymers and/or mixtures thereof.

4. A composition of claim 3 wherein said gluing agent is polyfurfurylalcohol.

5. A composition of claim 1 or 2 wherein said gluing agent is an inorganic binder selected from the group consisting of colloids, clays, concrete and cement components and/or mixtures thereof.

6. A composition of claim 5 wherein said gluing agent is a clay selected from the group comprising of sepiolite, bentonite, montmorillonite, LDH or pillared clays.

7. A composition of claim 5 or 6 wherein said gluing agent is a silica colloid selected from the group comprising of ludox^® or alumina colloids.

8. A composition of any of claims 5-7 wherein said gluing agent is a mixture of clay, for example sepiolite and ludox^®, which are preferably present in concentration by weight of about 30 to 70 % sepiolite and about 30 to 70 % ludox^®.

9. A composition according to any of the previous claims 1-8 wherein said active catalytic phase is MnO₂, preferably γ-MnO₂.

10. Use of a gluing agent in a composition as defined in any of claims 1-9.

11. A method for the generation of a composition of any of claims 1-9 wherein said active catalytic phase is deposited onto said support by bringing both active phase and support in contact with a gluing agent.

12. A method according to claim 11 wherein said binding agent is furfurylalcohol and/or polymers thereof undergoing a calcination step.

13. A method according to claim 11 or 12 wherein said calcination is a two-step calcination process.

5

14. A monolith consisting of a bilayer of which one layer has an unidirectional structure and the second layer is provided with a relief.

15. A monolith according to claim 14 wherein said relief structure consists of a bidirectional pattern.

0

16. A monolith according to claim 15 wherein said bidirectional pattern is made by folding the layer in two directions perpendicular onto each other.

17. A monolith according to claim 14, 15 or 16 whereon a composition according to any of claims 1- [5 9 is deposited.

18. Use of a γ-MnO₂ as a catalyst for removal of nitrogen-containing Volatile Organic Compounds (N-VOC's).

20 19. Use of a γ-MnO₂ glued on a support for removal of nitrogen-containing Volatile Organic Compounds (N-VOC's).

20. Use of a γ-MnO₂, preferably glued on a support, as a catalyst for the combined removal of nitrogen, sulphur- and/or halogen-containing Volatile Organic Compounds (N-, S- and halogen-

25 VOC's).

21. Use of a γ-MnO₂ glued on a monolith support according to any of claims 14-16.