Catalytic bromine recovery: An enabling technology for emerging alkane functionalization processes

Article abstract of DOI:10.1002/cctc.201300609

Making a quick recovery: The widespread implementation of bromination reactions to manufacture value-added products is contingent upon the development of sustainable and cost-effective means to recycle copious amounts of HBr byproduct. We report families of heterogeneous catalysts for the full recovery of Br₂ through HBr oxidation that display unprecedented low-temperature activity and stability. Copyright

Full text of DOI:10.1002/cctc.201300609

CHEMCATCHEM
COMMUNICATIONS
DOI: 10.1002/cctc.201300609
Catalytic Bromine Recovery: An Enabling Technology for
Emerging Alkane Functionalization Processes
Maximilian Moser, Laura Rodrꢀguez-Garcꢀa, Amol P. Amrute, and Javier Pꢁrez-Ramꢀrez*[a]
The functionalization of light alkanes into value-added prod-
ucts represents one of the most relevant and challenging areas
in catalysis research.[1] In particular, energy-efficient and cost-
effective processes to selectively convert the abundant re-
serves of natural gas into chemical intermediates and fuels are
highly sought. Pioneering work by Olah et al.[2] and more
recent studies by McFarland, Stucky, and co-workers[3,4] have
shown that the bromination of light alkanes to the corre-
sponding alkyl bromides followed by a catalyzed elimination
reaction offers an attractive route to obtain a broad spectrum
of desirable products (Figure 1). Bromine-mediated reactions of
of alkanes. A suitable halogen regeneration technology would
also aid the valorization of waste HBr streams originating from
the manufacture of organobromides employed as flame retard-
ants, polymers, and pharmaceutical intermediates.[5]
Bromine recovery can be achieved by HBr electrolysis,[6] HBr
oxidation,[5,7] and HBr absorption by a cataloreactant followed
by its reoxidation.[4a,b] To our knowledge, none of these pro-
cesses are established on a technical scale. The catalyzed gas-
phase oxidation of HBr with O2 or air (2HBr+1/2O2$Br2 +
H2O, DH0 =ꢀ138 kJmolꢀ1) is particularly attractive owing to its
low energy requirement and the relative simplicity of the pro-
cess. Nonetheless, the identification of highly active and stable
catalysts can be anticipated as critical owing to the exothermic
and corrosive nature of the reaction. Different materials have
been patented,[5] of which supported cerium-based com-
pounds are the most prominent.[7] However, the absence of
systematic studies aimed at screening the performance of po-
tential candidates and the limited understanding of the reac-
tion mechanism have hindered the design of suitable HBr oxi-
dation catalysts and their large-scale implementation. Herein,
we introduce several efficient heterogeneous catalysts for this
reaction that enable bromine recovery at relatively low tem-
peratures. These catalytic materials will improve the present
manufacture of organobromides by recycling the HBr byprod-
uct. Moreover, they can be regarded as a key component for
the practical realization of bromine-mediated alkane upgrading
processes.
Figure 1. The sustainable Br2-mediated conversion of light alkanes into val-
uable chemical intermediates and fuels requires an integrated HBr recycling
process to regenerate the halogen.
hydrocarbons are far more selective and occur under much
milder conditions (typically 475 K and 100–200 kPa) than classi-
cal alkane upgrading processes (steam reforming, steam crack-
ing, and dehydrogenation), which leads to increased product
yields, energy savings, and decreased CO2 emissions.[3] For in-
stance, the conversion of methane into olefins through methyl
bromide represents an intensified alternative to the conven-
tional methanol-based process by circumventing the costly in-
termediate syntheses of syngas and methanol.[4c] Another illus-
trative example comprises the bromine-based dehydrogena-
tion of propane, which gives a yield of propylene that is three
higher than the highest reported yields given by oxidative de-
hydrogenation.[4e] However, as shown in Figure 1, every mole
of alkane converted through this two-step process generates
two moles of HBr byproduct. Accordingly, the development of
a robust and economic process to recover Br2 from HBr is es-
sential to enable the sustainable bromine-mediated upgrading
Upon targeting the discovery of active and stable HBr oxida-
tion catalysts, it is inevitable to assimilate recent developments
in the analogous catalyzed oxidation of HCl (Deacon reaction)
for chlorine recovery in phosgenation processes.[8] Despite the
chemical similarity of these reactions, it is unclear whether
straightforward parallels can be drawn for catalyst and process
design. To shed light on this basic aspect, the oxidation of HBr
was investigated over representative Deacon catalysts. A
highly advantageous difference between the two reactions is
the absence of thermodynamic limitations in HBr oxidation
(Figure S1, Supporting Information), which a priori enables the
complete recovery of bromine. In HCl recycling, this can only
be attained by the costly combination of catalytic oxidation
and electrolysis processes.[8d] As shown in Figure 2, most of the
catalysts investigated exhibited significant HBr conversion in
the range of 400–600 K, which represents a decrease of 120–
330 K in the operating temperature window relative to that re-
quired for HCl oxidation for the same conversion level. The
case of U3O8/ZrO2, with a shift of approximately 330 K, is exem-
plary. As a result, the HBr oxidation performance of the least
active HCl oxidation catalyst is comparable to that of the most
active HBr oxidation catalyst, RuO2/TiO2. The performance of
[a] M. Moser, L. Rodrꢀguez-Garcꢀa, A. P. Amrute, Prof. J. Pꢁrez-Ramꢀrez
Institute for Chemical and Bioengineering
Department of Chemistry and Applied Biosciences ETH Zurich
Wolfgang-Pauli-Strasse 10, 8093 Zurich (Switzerland)
Fax: (+41)44-633-14-05
E-mail: jpr@chem.ethz.ch
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201300609.
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 3520 – 3523 3520

CHEMCATCHEM
COMMUNICATIONS
www.chemcatchem.org
*
)
Figure 3. The thermal activation of the impregnated RuCl3/TiO2 sample (
should be tuned to maximize the HBr oxidation performance of the result-
ing RuO2/TiO2 catalyst (measured at 433 K). The HAADF-STEM image clearly
shows the filmlike nanostructure of RuO2 on the TiO2-rutile particles. The
*
corresponding preparation with RuBr3/TiO2 ( ) resulted in a much less active
catalyst.
complete decomposition of the chloride below this tempera-
ture and the sintering of RuO2 above this temperature reduce
the activity. The catalyst calcined at the optimum temperature
consists of a fully decomposed oxide phase, which is highly
dispersed as thin layers formed by epitaxial growth on the sur-
face of the TiO2-rutile carrier [the high-angle annular dark field
(HAADF) scanning transmission electron microscopy (STEM)
image is given in Figure 3]. The decomposition temperature of
the chloride precursor is significantly higher than the tempera-
ture required for HBr oxidation. Considering the TGA results
(Figure S3, Supporting Information), it appears advantageous
to use RuBr3 as the precursor for the active RuO2 phase, which
enables the catalyst to be calcined at lower temperature, and
this leads to improved RuO2 dispersion. However, this was im-
practical owing to the low solubility of RuBr3 in water, which
led to inefficient impregnation and poor performance of the
Figure 2. Steady-state conversion of a) HBr and b) HCl over different cata-
lysts versus bed temperature. Other conditions are given in the Experimental
Section.
CeO2/ZrO2 was inferior to that of RuO2 and U3O8-based cata-
lysts. Cerium compounds have frequently been studied as HBr
oxidation catalysts.[5,7] However, the nature of the active phase
is unclear. Catalytic evaluation of the bulk phases enabled us
to conclude, in contrast to earlier claims,[7b] the inactivity of
CeBr3 (Figure S2, Supporting Information). The bromide only
became active upon decomposition to CeO2, that is, above
550 K [the thermogravimetric analysis (TGA) profile is given in
Figure S3, Supporting Information]. The absence of CeBr3 was
confirmed by XRD phase analysis of the used catalyst (Fig-
ure S2, Supporting Information). The same observations were
made for RuBr3, which becomes active upon oxidation to RuO2
over 550 K. However, IrBr3 was inactive up to 700 K owing to
its higher oxidation temperature (Figure S3, Supporting Infor-
mation). These results show that bulk halides are inactive for
HBr oxidation. Instead, the oxidic phase is required to catalyze
the reaction. This does not exclude the formation of surface
oxybromide species under reaction conditions analogous to
those under which an oxychloride phase was identified during
HCl oxidation over RuO2.[8a,b,e]
*
resulting sample (Figure 3, ).
The stark performance differences observed over the cata-
lysts is related to the mechanistic and kinetic differences be-
tween the two reactions. Toftelund et al.[9] conducted the first
theoretical study by using DFT to compare hydrogen halide
oxidations over defect-free rutile-type oxides and by selecting
the oxygen dissociation energy as the descriptor of catalytic
activity. In line with our experimental results, RuO2 was identi-
fied as the most active material in HBr oxidation. However, the
computational study also predicted that RuO2 would exhibit
similar activity in the oxidation of both HBr and HCl at high
coverage. This is the regime in which our catalytic tests were
conducted (ambient pressure and high partial pressures of re-
actants); we observed an equivalent HX conversion over RuO2/
TiO2 at a much lower temperature (by 120 K) in HBr oxidation
that in HCl oxidation (Figure 2). We confirmed the same differ-
ences upon evaluating the activity of bulk RuO2 (not shown).
Accordingly, the use of the oxygen dissociation energy as the
only determining parameter for the activity of these catalysts
may be a questionable approach. Early studies on the oxida-
tion of HCl demonstrated that the recombination of surface
The calcination conditions have a decisive effect on the cata-
lyst activity with respect to the degree of decomposition of
the metal halide precursor and to the dispersion of the active
phase. This aspect is demonstrated in Figure 3 for the rutheni-
um-based catalyst derived from the chloride precursor. HBr
conversion shows a distinct volcano plot in relation to the cal-
cination temperature and reaches a maximum at 723 K. The in-
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 3520 – 3523 3521

CHEMCATCHEM
COMMUNICATIONS
www.chemcatchem.org
atomic chlorine species is the most energy demanding
step.[8a,b] Therefore, the substantially lower operating tempera-
ture required for HBr oxidation relative to that required for HCl
oxidation over all of the catalysts is expected to be related to
the fact that the energy barrier for the recombination of bro-
mine is lower than that of chlorine. This should be associated
with the fact that the binding energy of bromides to an active
metal surface is weaker than that of chlorides to the same sur-
face. Furthermore, we show that a higher activity, evaluated by
the temperature needed to attain 20% conversion (T20), over
three of the supported catalysts investigated correlates with
a lower oxidative decomposition temperature of their metal
halides (Toxidation, Table 1). Thus, as bulk bromides are less stable
to the absence of thermodynamic limitations and the ease of
bromine evolution from the catalyst surface. The latter is relat-
ed to the fact that the surface bromide species are less stable
than the surface chloride species. This enables the attainment
of approximately 100% Br2 recovery at mild temperatures
(500–600 K). None of the above catalysts experience bulk bro-
mination under the reaction conditions, which secures the
stable performance. These catalysts can be applied to regener-
ate bromine from HBr waste streams produced in the manu-
facture of organobromides and will definitively aid the future
realization of sustainable bromine-mediated alkane functionali-
zation processes.
Experimental Section
Table S1 (Supporting Information) summarizes all the supported
catalysts evaluated in this work. In their preparation, the chloride,
bromide, and nitrate precursors of the active phases were dis-
solved in water according to the desired metal loading and subse-
quently applied to the carrier (rutile TiO2, monoclinic ZrO2, and
SiO2) by dry impregnation, followed by drying and calcination. The
gas-phase oxidation of HBr was studied at ambient pressure in
a continuous-flow fixed-bed reactor setup similar to that used for
HCl oxidation.[8g] The line downstream of the reactor was heated at
423 K to prevent the condensation of Br2. The catalyst (0.25 g, par-
ticle size=0.4–0.6 mm) was loaded in an 8 mm i.d. quartz micro-
reactor and pretreated under an atmosphere of N2 at 473 K for
30 min. Thereafter, a total flow of 166 cm3 STP minꢀ1 with 10 vol.%
HBr (or HCl) and 20 vol.% O2 balanced in N2 was fed to the reactor
at temperatures in the range of 400–823 K. Steady-state catalytic
data were collected after at least 1 h stabilization at each condi-
tion. The absence of diffusion limitations was secured by the fulfill-
ment of the Carberry and Weisz–Prater criteria. The amount of Br2
(or Cl2) at the reactor outlet was quantified by iodometric titration
by using a Mettler Toledo G20 compact titrator. The percentage of
HX (X=Br, Cl) conversion was determined according to Equa-
tion (1). After the test, the reactor was rapidly cooled to room tem-
perature, and the catalyst was recovered for characterization.
Table 1. Temperature required to achieve 20% conversion of the hydro-
gen halide over selected catalysts, T20, and oxidation temperature of the
bulk metal halides, Toxidation
.
[a]
[b]
Catalyst
HBr
T20 [K]
Toxidation [K]
HCl
Metal
Metal
oxidation
oxidation
bromide
chloride
RuO2/TiO2
CeO2/ZrO2
IrO2/TiO2
430
495
566
544
705
700
612
585
900
686
840
930
[a] Determined from Figure 2. [b] Determined from the derivative of the
weight loss profiles in Figure S3c (Supporting Information).
towards oxidation than chlorides, HBr oxidation operates at
a significantly lower temperature than HCl oxidation.
An important aspect for the practical realization of these cat-
alysts comprises their long-term durability under operating
conditions. We observed the stable performance of the materi-
als (with the exception of CuO/SiO2 and Cr2O3/SiO2) on the
basis of the identical conversion profiles obtained by following
multiple cycles in the temperature ramping tests shown in
Figure 2. These conclusions were verified by the stable HBr
conversion attained in isothermal tests up to 60 h over the
most active catalytic materials identified in this study, RuO2/
TiO2 (at 523 K) and U3O8/ZrO2 (at 563 K). The notable stability
of these catalysts, which did not experience any metal loss,
can be related to the fact that the samples retained their
oxidic active phase and crystalline structure upon use (Fig-
ure S4, Supporting Information). The lower temperatures re-
quired for HBr oxidation with respect to HCl oxidation favor
a higher stability in the former reaction. Bulk bromination of
the catalysts would induce fast deactivation by volatilization of
the halide phases. This is what happened in the case of CuO/
SiO2 and Cr2O3/SiO2 catalysts, which are also highly unstable in
HCl oxidation.[8f,g] Consequently, these two materials are the
least active catalysts in HBr oxidation (see Figure 2a and the
Table of Contents graphic).
2 mole X2 at reactor outlet
ð1Þ
XHX
¼
ꢁ 100
mole HX at reactor inlet
Keywords: alkane activation · bromine recovery · hydrogen
bromide · heterogeneous catalysis · oxidation
[1] a) A. E. Shilov, G. B. Shul’pin, Chem. Rev. 1997, 97, 2879–2932; b) F.
Cavani, N. Ballarini, A. Cericola, Catal. Today 2007, 127, 113–131; c) R. Pal-
kovits, M. Antonietti, P. Kuhn, A. Thomas, F. Schꢃth, Angew. Chem. 2009,
121, 7042–7045; Angew. Chem. Int. Ed. 2009, 48, 6909–6912; d) H.
Schwarz, Angew. Chem. 2011, 123, 10276–10297; Angew. Chem. Int. Ed.
2011, 50, 10096–10115.
[2] G. A. Olah, B. Gupta, M. Farina, J. D. Felberg, W. M. Ip, A. Husain, R. Kar-
peles, K. Lammertsma, A. K. Melhotra, N. J. Trivedi, J. Am. Chem. Soc.
1985, 107, 7097–7105.
[3] E. McFarland, Science 2012, 338, 340–342.
[4] a) X.-P. Zhou, A. Yilmaz, G. A. Yilmaz, I. M. Lorkovic, L. E. Laverman, M.
Weiss, J. H. Sherman, E. W. McFarland, G. D. Stucky, P. C. Ford, Chem.
Commun. 2003, 2294–2295; b) A. Breed, M. F. Doherty, S. Gadewar, P.
Grosso, I. M. Lorkovic, E. W. McFarland, M. J. Weiss, Catal. Today 2005,
106, 301–304; c) A. Zhang, S. Sun, Z. J. A. Komon, N. Osterwalder, S. Ga-
dewar, P. Stoimenov, D. J. Auerbach, G. D. Stucky, E. W. McFarland, Phys.
Chem. Chem. Phys. 2011, 13, 2550–2555; d) J. J. Waycuilis (Marathon GTF
In conclusion, we have demonstrated various supported cat-
alysts, based on RuO2, U3O8, and CeO2, which are suitable for
the oxidation of HBr to Br2 in a continuous flow reactor. Con-
siderably superior performances were obtained relative to
those obtained in the analogous oxidation of HCl to Cl2 owing
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 3520 – 3523 3522

CHEMCATCHEM
COMMUNICATIONS
www.chemcatchem.org
Technology Ltd.), US8232441B2, 2012; e) K. Ding, H. Metiu, G. D. Stucky,
ChemCatChem 2013, 5, 1906–1910.
[5] P. F. Schubert, R. D. Beatty, S. Mahajan, in Environmental Catalysis,
Vol. 552, ACS Symposium Series, American Chemical Society, Washington,
1994, pp. 405–419.
[6] a) J.-S. Park, C. Chen, N. L. Wieder, J. M. Vohs, R. J. Gorte, Electrochim. Acta
2011, 56, 1581–1584; b) A. Ivanovskaya, N. Singh, R.-F. Liu, H. Kreutzer, J.
Baltrusaitis, T. V. Nguyen, H. Metiu, E. McFarland, Langmuir 2013, 29,
480–492.
120, 2161–2164; Angew. Chem. Int. Ed. 2008, 47, 2131–2134; c) K. Seki,
Catal. Surv. Asia 2010, 14, 168–175; d) J. Pꢁrez-Ramꢀrez, C. Mondelli, T.
Schmidt, O. F.-K. Schlꢃter, A. Wolf, L. Mleczko, T. Dreier, Energy Environ.
Sci. 2011, 4, 4786–4799; e) D. Teschner, G. Novell-Leruth, R. Farra, A.
Knop-Gericke, R. Schlçgl, L. Szentmiklꢄsi, M. G. Hevia, H. Soerijanto, R.
Schomꢅcker, J. Pꢁrez-Ramꢀrez, N. Lꢄpez, Nat. Chem. 2012, 4, 739–745;
f) A. P. Amrute, C. Mondelli, J. Pꢁrez-Ramꢀrez, Catal. Sci. Technol. 2012, 2,
2057–2065; g) A. P. Amrute, F. Krumeich, C. Mondelli, J. Pꢁrez-Ramꢀrez,
Chem. Sci. 2013, 4, 2209–2217.
[7] a) M. Mugdan (The Distillers Company Ltd.), US2536457, 1951; b) P. Schu-
bert, D. W. Schubert, A. R. Smith, S. Mahajan, T. Rostrup-Nielsen (Catalyti-
ca Inc.), WO93/14868, 1993; c) J. T. Hall, G. B. Mckinnie (Albemarle Corpo-
ration), EP2117999A1, 2009.
[9] A. Toftelund, I. C. Man, H. A. Hansen, F. Abild-Pedersen, T. Bligaard, J.
Rossmeisl, F. Studt, ChemCatChem 2012, 4, 1856–1861.
[8] a) N. Lꢄpez, J. Gꢄmez-Segura, R. P. Marꢀn, J. Pꢁrez-Ramꢀrez, J. Catal. 2008,
255, 29–39; b) D. Crihan, M. Knapp, S. Zweidinger, E. Lundgren, C. J.
Weststrate, J. N. Andersen, A. P. Seitsonen, H. Over, Angew. Chem. 2008,
Received: July 25, 2013
Published online on September 9, 2013
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 3520 – 3523 3523

Copyright of ChemCatChem is the property of Wiley-Blackwell and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.