673

A publication of

CHEMICAL ENGINEERING TRANSACTIONS

The Italian Association
VOL. 32, 2013 of Chemical Engineering
Online at: www.aidic.it/cet
Chief Editors: Sauro Pierucci, Jiří J. Klemeš
Copyright © 2013, AIDIC Servizi S.r.l.,
ISBN 978-88-95608-23-5; ISSN 1974-9791

Hydrogen Peroxide Direct Synthesis: Enhancement of

Selectivity and Production with non-Conventional Methods
Pierdomenico Biasia,*, Juan Garcia Sernab, Tapio O. Salmia, Jyri-Pekka
Mikkolaa,c
a
Department of Chemical Engineering, Process Chemistry Centre (PCC), Laboratory of Industrial Chemistry and
Reaction Engineering Åbo Akademi University, Biskopsgatan 8, ÅBO-TURKU, 20500, Finlan
b
High Pressure Processes Group, University of Valladolid, Valladolid, Spain
c
Department of Chemistry, Chemical-Biochemical Centre (KBC), Technical Chemistry, Umeå University, UMEÅ, SE-
90187, Sweden
bpierdom@abo.fi

The present work is part of a comprehensive study on the direct synthesis of hydrogen peroxide in
different fields, from chemistry to chemical engineering. Working on the different fields of the direct
synthesis gave the possibility to look at the results and the challenges from different viewpoints. Here was
investigated one parameter that enhances the direct synthesis. The H2/Pd ratio is the key parameter that
has to be investigated and optimize to enhance the hydrogen peroxide direct synthesis. Two reactors were
built to study deeply the H2/Pd ratio and to demonstrate how this parameter can affect the direct synthesis
both in batch and continuous reactor with non-conventional experiments/methods. 1) A batch reactor was
utilized as a “starving reactor” to enhance the productivity of hydrogen peroxide and to try to keep constant
the selectivity. The starving method consists in refilling the hydrogen when it is consumed in the reactor. 2)
A trickle bed reactor was utilized with a gradient of catalyst along the reactor to maximize both production
and selectivity of hydrogen peroxide. The distribution of the catalyst along the bed gave the possibility to
significantly improve the selectivity and the production of hydrogen peroxide (up to 0.5% in selected
conditions). Higher production rate and selectivity were found when the catalyst concentration decreases
along the bed from the top to the bottom compared to the uniformly dispersed catalyst. Selectivity in the
batch reactor was enhanced by 5% and in the continuous reactor of 10%. The non-conventional
experimental methods have been found to be novelty concepts to enhance the hydrogen peroxide direct
synthesis.

1. Introduction
Hydrogen peroxide is considered an inherently green oxidant, one of the cleanest, most versatile chemical
oxidants available, since it decomposes to yield only water and oxygen as the reaction products. H2O2 is
also one of the most efficient oxidizing agents by virtue of its high active oxygen content (about 47%), only
next to molecular oxygen.
Hydrogen peroxide was discovered by the world-famous French chemist Louis-Jacques Thénard in 1818.
Since then to the beginning of the 20th century, 3% H2O2 solutions were obtained from the hydrolysis of
barium peroxide with sulphuric acid. Several alternative processes, such as electrochemical and chemical
(oxidation of isopropanol or methylbenzyl alcohol, or antraquinone oxidation) processes have been
developed and are available for the production of H2O2.
Today, hydrogen peroxide is manufactured almost exclusively by the Riedl-Pfleiderer process and the
Hydrogen peroxide annual production is nowadays around three million tons and the growth of its market
is estimated around 4%/year (Arkema Website).
The main uses of hydrogen peroxide in europe are summarized below in the Figure 1 below. As revealed
by the Figure 1, the major part of the hydrogen peroxide, more than 80% is used for paper bleaching and
chemical synthesis.
674

Figure 1: Uses of hydrogen peroxide in Europe.

1.1 Why direct synthesis?

The "Strategic Research Agenda" (SRA) of the European Technology Platform for Sustainable Chemistry
reports that the direct synthesis of the H2O2 from H2/O2 and its integrated use in chemical processes is one
of the priorities for a sustainable chemistry (Centi, 2009). H2O2 is an ideal clean reagent, but its industrial
use is inhibited from the high cost of the H2O2. This is reflected by the complains of the H2O2 users, which
suffer rising in prices due to a market strictly constrained by few large producers (Centi, 2009).
The direct synthesis from H2/O2 with catalysts based on Pd allows to decrease more than half the cost of
commercial H2O2, making economically feasible the following processes: direct synthesis of phenol,
propene oxide, caprolactam. Moreover, the following industries will also benefit from the direct synthesis:
paper industry (as a bleaching, chlorine-free chemical), textile industry (as a bleaching agent replacing
sodium chlorite and hypochlorite or a component in many detergents), chemical industry itself (in
oxidations, epoxydations, hydroxylation, as an initiator in polymerisations) and in the electronic industry (to
clean semiconductors). Other potential applications of the direct synthesis are primary and waste water
treatment (as a disinfectant or sulphite odour control), several inorganic syntheses (manufacture of sodium
percarbonate and sodium perborate, used as mild bleaches in laundry detergents) and several domestic
use (disinfectant, bleaching of hairs and tooth whitener, detergents).
Most of the applications mentioned call for a capillary network of distribution, in relatively limited individual
quantities. This fragmentation of usage, together with the hazard of transportation, suggests another
breaking observation: developing small-scale production units would be a much required, a further goal for
a new process. Again, this is in conflict with the actual situation where almost universally H2O2 is produced
through a complex process, with a critical environmental impact, which is economically feasible only at
large scale. Consequently, the H2O2 production is presently concentrated in a few large plants and little
space for competition is left to small and medium producers (Campos-Martin et al., 2006), unless a new,
economically viable technology emerges (Centi et al., 2009).
The H2O2 effectiveness as an oxidant is also due to the large abundance of oxygen in its molecule (47%),
where only pure O2 has a higher value. In addition, use of H2O2 produces harmless water and oxygen, as
required by a green chemical.
The arguments above lead to the choice of investigating the process of direct synthesis of H2O2 from H2
and O2. The reaction is apparently extremely simple, but unfortunately side reactions are likely to occur,
partly connected with the reactivity of H2O2, that causes hazard in manipulation and transportation, but it is
also the reason for its countless applications, partly because they involve water, an extremely stable
molecule, as product.

The technology related to the direct synthesis will benefit the industries lowering the price of hydrogen
peroxide and reducing the wastes related to the auto-oxidation process. The direct synthesis of hydrogen
peroxide is the most promising alternative to the auto-oxidation process. Table 1 compares the benefits
and drawbacks of the auto-oxidation process and the direct synthesis. The direct synthesis is still not well
enough developed to be considered as an alternative to the auto-oxidation process, but research efforts
will be helpful to solve all the issues related to the selectivity and safety issues of the direct synthesis
process.
 675

Table 1: Comparison between auto-oxidation process and direct synthesis of hydrogen peroxide.
PROCESS Indirect synthesis Direct synthesis
auto-oxidation process
PRINCIPLE Sequential hydrogenation and H2+O2=H2O2
oxidation of organic molecules,
release of H2O2 and recycle of the
organic molecules
GENERAL FEATURE A very famous, well-known and A very simple process but
complex process developed during a still lack of maturity
very long time
CATALYST Pd in the hydrogenation step Pd- and Au-based catalysts
REACTION Working solution of a mixture of H2O, MeOH or EtOH,
ENVIRONMENT organic molecules CO2 used as a co-solvent
REACTOR SYSTEM A complex system that counts of Synthesis done in one
numerous and sequential big reactors reactor
SELECTIVITY High It is an issue
SAFETY Safe Can be made safe
ON SITE PRODUCTION Impossible Possible

1.2 Scope of the research on the H2O2 direct synthesis

The present research is part of the research strategy on the hydrogen peroxide direct synthesis. The
combination of catalysis and chemical engineering applied to the hydrogen peroxide direct synthesis is
fundamental to understand this complicate reaction. The general aim of the work is to demonstrate that is
necessary to understand both the catalytic and reactor effects on the direct synthesis (Biasi et al., 2012a,
Gemo et al., 2012; Kilpio et al., 2012). The development of new catalysts has to be investigated coupled to
reactor development in order to find the most suitable conditions for high production rate and selectivity
towards hydrogen peroxide (Biasi et al., 2010; Biasi et al., 2011a; Biasi et al., 2011b; Biasi et al., 2012b).
The all work was structured as reported in the research diagram flow-sheet in Figure 2.

Catalyst Development Trickle Bed Reactor

&
Reaction Engineering Optimization

Batch Reactor Kinetic Kinetic Studies Kinetic Studies Trickle Bed Reactor
In In
Modelling Modelling Batch Reactor Trickle Bed Reactor Modelling

Residence Time
Distribution
Studies
Gas Solubility Studies
&
Modelling

Figure 2: Research diagram flow-sheet for hydrogen peroxide direct synthesis

Here we investigate the H2/Pd ratio a parameter of particular importance that has been found to be of
critical importance in the direct synthesis both in batch and in continuous reactors. H2/Pd ratio was
investigated with new concepts in reactor design like the “starving reactor” and the “catalyst gradient
distribution continuous reactor”.
676

2. Experimental Set-Up

2.1 Batch Reactor

The experiments were performed in a tailored batch reactor, i.e., a stainless steel autoclave reactor (Parr)
with a volume of 600 mL and a maximum working pressure of 200 bar. The mixing was performed by a
Heidolph RZR 2021 rotor operating at 1000 rpm. The reactor was fitted with a K-type thermocouple
connected to a PC that allowed acquisition of temperature data during the experiments. CO2 and O2 were
loaded into the reactor directly from gas cylinders. Hydrogen was fed by a Brooks mass flow controller to
avoid overpressure in the reactor and to ensure a constant and reproducible flow. A 35 mL vessel coupled
with a pressure transducer and a heating jacket was placed before the mass flow controller (MFC) for
hydrogen feeding. This was used to calculate precisely the amount of hydrogen fed into the reactor.
Before the experimental programme was started, a 4 hour passivation treatment was performed on the
reactor, using 30 wt.% nitric acid at 40 °C. Methanol was chosen as reactant medium.
The protocol developed for the experiments was as follows: 0.15 g of a 5% Pd/C (supplied by Degussa)
catalyst were loaded in the reactor. Carbon dioxide (18.4 bar) and oxygen (6 bar) were loaded into the
reactor directly from the cylinders at 25 °C, followed by the injection of 400 mL of methanol by an HPLC
pump. After both pressure and temperature were stable hydrogen was fed from the vessel into the reactor
through the mass flow controller. Three different experiments were conducted: 1) synthesis of hydrogen
peroxide with one hydrogen filling at the beginning of the reaction (Biasi et al., 2012b); 2) synthesis of
hydrogen peroxide with four hydrogen fillings along the reaction, after the totality of the hydrogen dissolved
in the liquid phase was completely consumed and 3) synthesis of hydrogen peroxide with four hydrogen
fillings, as soon as the hydrogen concentration in the liquid phase started to decrease.
2.2 Trickle Bed Reactor

The system consisted of four sections, liquid inlet, gas inlet, three-phase reaction, and biphasic separation.
The liquid was pumped at room temperature using a HPLC pump. The gases were fed independently from
gas cylinders (CO2/H2 95/5 mol% and with pure O2) controlled by mass flow controllers (MFC). The TBR
was made of AISI 316 stainless steel, 60 cm long and 1.5 cm I.D. (ca. 105 mL) passivated using nitric acid
30 % wt. during 4 h to prevent H2O2 decomposition due to iron catalysis. The bed is composed of 2 cm of
glass wool at the bottom and up to 110 g of quartz sand mixed with the catalyst (ca. 80-100 mL) reactor
can accommodate a catalyst bed up to 50 cm long. The reactor operated in continuous mode, with
cocurrent downflow gas and liquid phases. Liquid flow rate: 1 mLmin water (enhancers: 0.0003 M NaBr+
0.005 M H3PO4) and gas flow rate: 4 mL/min.
Water was used instead of methanol in the continuous fixed bed reactor. Water was chosen because the
hydrogen solubility is lower than in methanol. If methanol is used as liquid phase there is a risk that the
amount of H2 available for the direct synthesis is too high in the liquid phase, resulting in a fast H2
consumption. Since the aim of the study was to improve the reaction conditions water was chosen as
reactant medium to dose the hydrogen in the liquid phase in all the reactor length.

3. Results and Discussion

3.1 Batch Reactor

The experiments in the batch reactor were performed in order to find a relationship between the selectivity,
the production rate and the amount of hydrogen fed into the reactor (Figures 3 and 4). The same amount
of hydrogen was fed into the reactor straight after the gas dissolved in the liquid was completely
consumed. Both the hydrogen peroxide and water concentrations increased during the reaction. The
selectivity had different behaviour though: every time that fresh hydrogen was fed into the reactor, the
selectivity towards hydrogen peroxide increased; when hydrogen started to decrease, such selectivity
decreased as well. The results prove that the amount of hydrogen plays an important role in the direct
synthesis.
Regarding the importance of the H2/Pd ratio, experiments were conducted avoiding the complete depletion
of hydrogen in the liquid phase. In that sense, equal amounts of hydrogen were reloaded into the reactor a
certain time after the H2 peak was recorded with the Fugatron instrument. Such subsequent hydrogen
feeding policy demonstrated how a higher selectivity could be achieved compared to the former
experiments when the hydrogen was refilled after its complete consumption. The selectivity towards
hydrogen peroxide in this case reached a peak around 33 %, compared to only 27 % in the previous case,
677
678

mL of quartz sand and from the bottom the quantity of the catalyst was 12.5, 25 and 37.5 mg of catalyst in
each part of the quartz sand. The three different bed dispositions are reported in Figure 5.
From those preliminary results from the bed A) to the bed C) (Figure 5) there was an average
enhancement of 10 % in the selectivity and a 20 % of the H2O2 produced only with a different disposition of
the catalyst bed. The best achievement was found at 5 °C where the yield was improved by a value of 11
% from a 50 % with the bed A) to 60 % with the bed C). the production rate was 0.4 wt.% with the bed A)
and 0.5 wt.% with the bed C). the amount of the active catalyst was the same and if the hydrogenation
(H2+H2O2 over Pd catalyst to give 2 H2O) was not influenced by the H2/Pd ratio the results should have
been exactly the same. These results showed that the hypothesis made was correct because with this
different disposition the hydrogenation is kept lower in the bed C) then in the bed A). All the results made
with the bed B) were following the same trend than in the former cases decreasing the production and the
yield of H2O2 when the temperature is increased from 5 to 35 °C.

4. Conclusions
Different policies of H2 feeding gave the possibility to enhance the hydrogen peroxide production rate and
selectivity in a batch reactor. An improved selectivity was achieved by refilling hydrogen before its
complete consumption. These results gave further direction of the investigation on the role of hydrogen
dissolved in the direct synthesis. The investigation on the batch reactor was fundamental to program the
research activities in a continuous reactor.
A new trickle bed reactor was constructed with the information from the batch experiments. Following the
experiments that showed the importance of the H2/Pd ratio, the concept of “catalyst gradient distribution
continuous reactor” was introduced. The bed with a catalyst gradient was found to be very essential to
increase selectivity and productivity in a continuous reactor. These results confirmed that if the H2/Pd ratio
goes under some values, hydrogenation takes place as the main reaction.
Finally, the non-conventional methods to operate a reactor as the “starving reactor” and the “catalyst
gradient distribution continuous reactor” were a success to enhance the hydrogen peroxide direct
synthesis. These new concepts applied to reaction engineering and catalysis engineering will not benefit
only the direct synthesis, but also other reactions were selectivity is a really important issue.

References

Arkema Website http://www.arkema.com/sites/group/en/products/spotlight/h2o2.page, 11.01.2013

Biasi P., Menegazzo F., Pinna F., Eranen K., Canu P., Salmi T.O., 2010, Hydrogen Peroxide Direct
Synthesis: Selectivity Enhancement in a Trickle Bed Reactor, Industrial & Engineering Chemistry
Research, 49, 10627-10632.
Biasi P., Zancanella S., Pinna F., Canu P., Salmi T.O., 2011a, Hydrogen Peroxide Direct Synthesis : from
Catalyst preparation to Continuous Reactors, Chemical Engineering Transactions, 24, 49-54 DOI:
10.3303/CET1124009
Biasi P., Menegazzo F., Pinna F., Eranen K., Salmi T.O., Canu P., 2011b, Continuous H2O2 direct
synthesis over PdAu catalysts, Chemical Engineering Journal, 176-177, 172-177.
Biasi P., Gemo N., Hernandez Carucci J.R., Eranen K., Canu P., Salmi T.O., 2012a, Kinetics and
Mechanism of H2O2 Direct Synthesis over a Pd/C Catalyst in a Batch Reactor, Industrial & Engineering
Chemistry Research, 51, 8903-8912.
Biasi P., Canu P., Menegazzo F., Pinna F., Salmi T.O., 2012b, Direct Synthesis of Hydrogen Peroxide in a
Trickle Bed Reactor: Comparison of Pd-Based Catalysts, Industrial & Engineering Chemistry
Research, 51, 8883-8890.
Campos-Martin J.M., Blanco-Brieva G., Fierro J.L.G., 2006, Hydrogen Peroxide Synthesis: An Outlook
beyond the Anthraquinone Process, Angewandte Chemie International Edition, 45, 6962–6984.
Centi G., Perathoner S., Abate S., 2009, Direct synthesis of hydrogen peroxide: recent advances, Modern
Heterogeneous Oxidation Catalysis, Wiley-VCH Weinheim, Germany, pp. 253-287.
Gemo N., Biasi P., Salmi T.O., Canu P., 2012, H2 solubility in methanol in the presence of CO2 and O2,
The Journal of Chemical Thermodynamics, 54, 1-9.
Kilpio T., Biasi P., Bittante A., Salmi T., Warna J., 2012, Modeling of Direct Synthesis of Hydrogen
Peroxide in a Packed-Bed Reactor, Industrial & Engineering Chemistry Research, 51, 13366-13378.