Applications of Hydrogen Peroxide and Derivatives

By Craig W. Jones

The Royal Society of Chemistry

Copyright © 1999 The Royal Society of Chemistry
All rights reserved.
ISBN: 978-0-85404-536-5

Contents

Chapter 1 Introduction to the Preparation and Properties of Hydrogen Peroxide, 1,
Chapter 2 Activation of Hydrogen Peroxide Using Inorganic and Organic Species, 37,
Chapter 3 Application of Hydrogen Peroxide for the Synthesis of Fine Chemicals, 79,
Chapter 4 Heterogeneous Activation and Application of Hydrogen Peroxide, 179,
Chapter 5 Environmental Applications of Hydrogen Peroxide, 207,
Chapter 6 Miscellaneous Uses for Hydrogen Peroxide Technology, 231,
Subject Index, 257,

CHAPTER 1

Introduction to the Preparation and Properties of Hydrogen Peroxide

1 Introduction

The following chapter will discuss the preparation of hydrogen peroxide, historically, the present day and future vistas for its in situ preparation. A brief introduction to the physical properties of hydrogen peroxide will also be made for the sake of completeness. Finally, the chapter will conclude with a practical approach to the safe handling of peroxygen species, destruction of residual peroxygens, and the toxicological and occupational health considerations required when handling hydrogen peroxide.

2 Industrial Manufacture of Hydrogen Peroxide

The industrial manufacture of hydrogen peroxide can be traced back to its isolation in 1818 by L. J. Thenard. Thenard reacted barium peroxide with nitric acid to produce a low concentration of aqueous hydrogen peroxide; the process can, however, be significantly improved by the use of hydrochloric acid. The hydrogen peroxide is formed in conjunction with barium chloride, both of which are soluble in water. The barium chloride is subsequently removed by precipitation with sulfuric acid (Figure 1.1).

Hence, Thenard gave birth to the first commercial manufacture of aqueous hydrogen peroxide, although it took over sixty years before Thenard’s wet chemical process was employed in a commercial capacity. The industrial production of hydrogen peroxide using the above route was still operating until the middle of the 20th century. At the turn of the 19th century, approximately 10 000 metric tonnes per annurn of barium peroxide were converted to about 2000 metric tonnes of hydrogen peroxide. Thenard’s process has, however, some major drawbacks which quenched the expectant explosion of its use in an aqueous form. Firstly, only three percent m/m aqueous hydrogen peroxide solutions were manufactured using the barium peroxide process, and hence only a limited market was afforded because production costs were prohibitively high. Further, due to the high levels of impurities present in the isolated hydrogen peroxide, subsequent stability was poor.

The disadvantages of the process discovered by Thenard were largely alleviated by the discovery in 1853 by Meidinger that hydrogen peroxide could be formed electrolytically from aqueous sulfuric acid. Berthelot later showed that peroxodisulfuric acid was the intermediate formed, which was subsequently hydrolysed to hydrogen peroxide, and sulfuric acid (Figure 1.2).

The first hydrogen peroxide plant to go on-stream based on the electrochemical process was in 1908 at the Österreichische Chemische Werke in Weissenstein. The Weissenstein process was adapted in 1910 to afford the Müncher process developed by Pietzsch and Adolph at the Elecktrochemische Werke, Munich. In 1924, Reidel and Lowenstein used ammonium sulfate under the conditions of electrolysis instead of sulfuric acid, and the resulting ammonium peroxodisulfate (Reidel–Lowenstein process) or potassium peroxo-disulfate (Pietzsch–Adolph process) was hydrolysed to hydrogen peroxide. As a result of this process, production of hydrogen peroxide as 100% m/m rose to approximately 35 000 metric tonnes per annum.

In 1901, Manchot made a decisive breakthrough in the industrial preparation of hydrogen peroxide. Manchot observed that autoxidizable compounds like hydroquinones or hydrazobenzenes react quantitatively under alkaline conditions to form peroxides. In 1932, Walton and Filson proposed to produce hydrogen peroxide via alternating oxidation and reduction of hydrazo-benzenes. Subsequently, Pfleiderer developed a process for the alkaline autoxidation of hydrazobenzenes in which sodium peroxide was obtained, and sodium amalgam was used to reduce the azobenzene. A commercial plant based on this technology was operated by Kymmene AB in Kuisankoski, Finland.

The major drawbacks associated with the azobenzene process, i.e. hydrogenation of azobenzene with sodium amalgam, and oxidation of hydrazobenzene in alkaline solution, were ultimately resolved by Riedl. Riedl employed polynuclear hydroquinones. Based on Reidl and Pfleiderer’s work, BASF developed, between 1935 and 1945, the anthroquinone process (often referred to as the AO process) in a pilot plant with a monthly production of 30 metric tonnes. Two large plants were then constructed at Heidebreck and Waldenberg, each having a capacity of 2000 metric tonnes per annum. Both plants were partially complete when construction was halted at the end of World War Two. In 1953, E.I. Dupont de Nemours commissioned the first hydrogen peroxide plant using the AO process, and consequently the production capacity of hydrogen peroxide was greatly increased. In 1996, world capacity stood at 1.3 x 106 metric tonnes as 100% m/m hydrogen peroxide.

The underlying chemistry of the AO process is outlined in Figure 1.3 and a typical autoxidation plant schematic is summarized in Figure 1.4.

The features of all AO processes remain basically the same, and can be described as follows. A 2-alkylanthraquinone is dissolved in a suitable solvent or solvent mixture which is catalytically hydrogenated to the corresponding 2-alkylanthrahydroquinone. The 2-alkylanthraquinone solution is commonly referred to as the reaction carrier, hydrogen carrier or working material. The 2-alkylanthraquinone–solvent mixture is called the working solution. Carriers employed industrially include 2-tert-amylanthraquinone, 2-iso-sec-amylanthraquinone and 2-ethylanthraquinone. The working solution containing the carrier product alkylanthrahydroquinone is separated from the hydrogenation catalyst, and aerated with an oxygen-containing gas, nominally compressed air, to reform the alkylanthraquinone, and simultaneously forming hydrogen peroxide. The hydrogen peroxide is then extracted from the oxidized working solution using demineralized water, and the aqueous extract is then purified and concentrated by fractionation to the desired strength. The AO process, therefore, leads to the net formation of hydrogen peroxide from gaseous hydrogen and oxygen.

The choice of the quinone must be carefully made to ensure that the following criteria are optimized: good solubility of the quinone form, good solubility of the hydroquinone form, good resistance to non-specific oxidation and easy availability. The formation of degradation products, and their ability to be regenerated to active quinones also plays a rôle in the decision. A number of byproducts can be formed during the hydrogenation step, and these are summarized in Figure 1.5. The process when first engaged, contains in the working solution only the 2-a1kylanthraquinone species. The 2-a1kylanthraquinone forms a complex with the hydrogenation catalyst, which is usually a palladium metal. The complex then reacts with hydrogen to form a species now containing the metal and the 2-alkylhydroanthraquinone. The 2-alkylhydroanthraquinone is subject to a number of secondary reactions which are continuously taking place during each process cycle.

The 2-alkylhydroanthraquinone (A) when in contact with the catalyst will undergo a small amount of catalytic reduction (B) on the ring, initially on the unsubstituted ring, yielding a tetrahydroalkylanthrahydroquinone. Unfortunately, once the octa-product (C) is formed, it remains until purged owing to its very low rate of oxidation. Tautomerism of the 2-alkylhydroanthraquinone yields hydroxyanthrones (D, E) which can be further reduced to the anthrones (G, H). The epoxide (F) formed from the alkylhydroanthraquinone does not participate in the formation of hydrogen peroxide, and leads to a loss of active quinone. Measures have, therefore, been suggested for regenerating the tetra-hydro compound from the epoxide.

A number of additional processes are also required to maintain the AO process. For example, in order for the hydrogenation phase to run efficiently, part of the catalyst load is removed, regenerated and returned to the hydrogenator. The hydrogenation step is possibly the most important feature of the modern A0 process. Quinone decomposition products that cannot be regenerated into active quinones are always formed during the hydrogenation phase. Therefore a tremendous amount of effort has been invested in the development of new hydrogenation catalysts and hydrogenator designs which have, in some cases, deviated dramatically from the BASF principle. The hydrogenation step in the BASF plant (Figure 1.6) employs a Raney nickel catalyst at a slight excess of pressure. However, because Raney nickel is sensitive to oxygen, the working solution from the extraction, drying and purification steps cannot be fed directly into the hydrogenator. The working solution at this stage still contains residual hydrogen peroxide, and has to be decomposed over a supported Ni–Ag catalyst (Figure 1.7), together with a small amount of hydrogenated working solution (which also contains 2-alkylhydroanthraquinone). Such a step removes the hydrogen peroxide completely, thus extending the life of the Raney nickel catalyst.

The problem with Raney nickel as the hydrogenation catalyst is that it has a limited selectivity, i.e. the ratio of hydroquinone formation to the tetrahydro compound is low. BASF have largely alleviated this problem via pre-treatment of the catalyst with ammonium formate. The pyrophoric properties of Raney nickel also require more stringent safety procedures when handling the material. Despite the drawbacks of Raney nickel, the catalyst is still used in some AO plants. The majority of AO plants worldwide prefer, however, to employ palladium hydrogenation catalysts because of their higher selectivity, their greater stability towards hydrogen peroxide residues and the simplified handling procedures in comparison to the Raney nickel systems. Degussa have employed palladium black as the hydrogenation catalyst in the majority of their plants. The main advantages of the Degussa hydrogenation stage are: near-quantitative conversion of hydrogen, easy exchange of palladium black, the catalyst is non-pyrophoric and the palladium black is easily re-activated. Laporte chemicals made a significant breakthrough in the operation of the hydrogenation phase by employing supported palladium, which has a particle size diameter of 0.06–0.15 mm. The supported palladium catalyst allows for easier filtration, and recirculation of the catalyst back to the hydrogenator. Laporte, at the same time, also employed a new design for running the hydrogenation phase. Figure 1.8 illustrates the Laporte design.

The Laporte hydrogenator contains a series of tubes which dip just below the surface of the liquid. Hydrogen is then fed into the bottom of each tube, and small gas bubbles are formed. A counter current flow is set up due to the density difference between the solutions in the tube and the reactor. The palladium catalyst suspension is drawn into the tubes by a continuous movement of the working solution.

The problem with all three methods thus far discussed is the fact that the hydrogenator catalyst has to be removed prior to the formation of hydrogen peroxide. If the catalyst is not removed, then catastrophic dismutation of the hydrogen peroxide can occur. In response to the problem, FMC developed a mixed-bed hydrogenation process. The bed is impregnated with palladium, and hence the problem associated with catalyst removal is alleviated.

On an industrial scale, the catalyst-free hydrogenated working solution is generally oxidized with slight pressures of air (up to 0.5 MPa). The oxidation phase must satisfy several criteria, mainly economically driven, which include: small reactor volume to lower investment costs for equipment; efficient utilization of oxygen to reduce the volume of off-gas; and low compressor pressure to decrease energy costs. Like the hydrogenation phase, several companies have developed and used their own oxidation regimes. For example, BASF flow hydrogenated working solution through four oxidation columns arranged in series (Figure 1.9) as a cascade. The oxidized working solution then flows into an extractor tank. The nitrogen–oxygen mixture is compressed and fed into each of the four reactors.

Solvay Interox’s plant based at Warrington in the UK operates a co-current oxidation in a column. The whole volume of the reactor is used for air gassing (Figure 1.10). The air and hydrogenated working solution leave the top of the column and are fed into a separator. The air then reaches the two-stage activated carbon filters, which remove residual working solution and impurities. The working solution then passes to the extraction phase.

Finally, it is worth mentioning that Allied Colloids have employed a counter-flow oxidation reactor, which has a residence time of hydrogenated working solution of less than 2.5 min at a partial oxygen pressure of 70–100 kPa.

Inevitably, due to the constant circulation of working solution, by-products are formed from the working solution and the solvents. The by-products have to be purged from the system to prevent destabilization of the crude hydrogen peroxide, and an increase in density and viscosity of the working solution. Further, the impurities in the working solution cause a decrease in the surface tension, and encourage the formation of an emulsion, which can be difficult to destabilize. By-product formation can also cause deactivation of the hydro-genation catalyst, hence the working solution can be purified by a range of techniques which include treatment with alkaline solution, treatment with active aluminium oxide or magnesium oxide at about 150 °C, use of alkaline hydroxide such as calcium hydroxide, ammonia or amines in the presence of oxygen or hydrogen peroxide and treatment with sulfuric acid.

The crude hydrogen peroxide exiting the extraction phase requires purification. A number of methods have been devised for the treatment of crude hydrogen peroxide including the use of polyethylene, ion-exchangers and the use of hydrocarbon solvents. The purified hydrogen peroxide is then fed to a distillation column where it is concentrated to the usual commercial concentration range of 35–70% m/m. Solvay Interox produce 85% m/m hydrogen peroxide, but only use it captively for the preparation of 38% m/m peracetic acid used for the oxidation of cyclohexanone to ε-caprolactone. Higher strengths can be achieved as hydrogen peroxide does not form an azeotrope with water, but a number of technical safety requirements must be observed.

Before we leave the discussion of industrial processes, it is worth mentioning one other autoxidation process, based on the oxidation of propan-2-0l, developed by Shell Chemicals. The process was employed by Shell in its 15 000 metric tonnes per annum facility at Norco between 1957 and 1980. The process was discovered in 1954 by Harris, who showed that the oxidation of primary and secondary alcohols formed hydrogen peroxide, and the corresponding aldehyde or ketone (Figure 1.11).

Only propan-2-ol has had any industrial use since the aldehydes formed in the reaction with primary alcohols are easily oxidized. The oxidation of propan-2-ol in the liquid phase with oxygen does not require a special catalyst, because it is catalysed by a small amount of hydrogen peroxide, which is added to the feed-stream of the propan-2-ol in order to shorten the induction phase (Figure 1.12).

Reduction of by-products can be achieved by only partially oxidizing the propan-2-ol, and by carrying out the oxidation in several consecutive steps, at decreasing temperatures. The hydrogen peroxide yield is typically 90–94% with respect to the propan-2-ol, and the acetone yield is 92–94%.

Over the years, there have been many other methods proposed for the preparation and subsequent purification of hydrogen peroxide. However, to date no industrial plants have been designed and commissioned based on such technologies. For example, Arco have devised a method for the preparation of hydrogen peroxide based on the autoxidation of methyl benzyl alcohol isomers with molecular oxygen. The process employs ethylbenzene and water to extract the hydrogen peroxide from a mixture of methyl benzyl alcohol and other oxidation by-products. For safety reasons, the water is supplied as a downward-flowing stream in the reactor, together with an upward flow of ethylbenzene. The process also contains one further feature worthy of note, which is that the crude aqueous hydrogen peroxide is passed through a cross-linked polystyrene resin which has a macro-reticular structure. This resin purification step has the advantage that subsequent concentration stages are inherently safer due to the lower organic contents. A number of novel electrochemical processes for hydrogen insertion reactions into molecules have also been applied to the preparation of hydrogen peroxide. One process worth describing involves the electrochemical production of hydrogen peroxide together with the simultaneous production of ozone. The preparation of ozone is from the anode and of hydrogen peroxide from the cathode. The oxidants are generated from water and oxygen in a proton-exchange membrane (PEM) reactor. The optimum conditions for generating the oxidants were found by the workers to be a function of applied voltage, electrode materials, catalyst loadings, reactant flow-rates and pressure. The ozone is generated at room temperature and pressure using lead dioxide powder bonded to a proton exchange membrane (Nafion® 117). The maximum concentration of the ozone formed is about 3 mg dm-3 in the aqueous phase. The cathodic reaction during the preparation of the ozone is hydrogen, which is oxidized with oxygen at 15 psi and a flow-rate of 100 ml min-1. The electrocatalysts investigated were various loadings of gold, carbon and graphite powders which are bonded to the membrane or to a carbon fibre paper pressed against the membrane. Hydrogen peroxide was evolved from all the catalysts studied, with the graphite powders yielding the highest concentration (25 mg dm-3). This process may have potential for the destruction of low concentrations of hazardous organic compounds in water courses.

(Continues…)Excerpted from Applications of Hydrogen Peroxide and Derivatives by Craig W. Jones. Copyright © 1999 The Royal Society of Chemistry. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.