Version of Record: https://www.sciencedirect.

com/science/article/pii/S1566736718302541
Manuscript_295b2d7bb9143a34c22507ad09a31c20

Ambient pressure synthesis of ammonia using a microwave reactor

Christina Wildfirea,c*, Victor Abdelsayeda,b, Dushyant Shekhawata , Michael J. Spencera,b
a National Energy Technology Laboratory, Morgantown, WV 26507
b AECOM, Morgantown, WV 26506
c Oak Ridge Institute of Science and Education, Oak Ridge, TN 37830
*corresponding author: Christina.wildfire@netl.doe.gov

Abstract

Ammonia synthesis processes that can operate at low temperatures and pressure and have ability

for small-scale distributed production are of interest. Microwave assisted synthesis has shown

improvements in kinetics and does not have economy of scale limitations of the Haber-Bosch

process. A microwave reactor was used to produce ammonia at temperatures from 300-500°C.

Three catalyst systems were compared: commercial Fe catalyst, and supported Ru/Al2O3, and

Fe/Al2O3 catalysts. The ammonia production of the dispersed catalysts was significantly higher

than the commercial catalyst with maximum production of 128 µmol/g/h at 300°C. This increase

in production is attributed to an optimal electron movement under the microwave field.

1. Introduction

Ammonia is the second most abundant synthetic chemical produced today mainly by the

Haber-Bosch process and consumes between 1- 2% of the total energy supply in the world [1].

This synthesis reaction is typically carried out at temperatures between 400-500°C, pressures

above 100 bar and is efficient at large scale [2, 3] . These large-scale plants require considerable

energy and water resources and a distribution infrastructure to get the products to consumers. If

the ammonia synthesis could be scaled down the plants could be located more remotely for

instance closer to crop sites and could take advantage of renewable energy resources. The ideal

© 2018 published by Elsevier. This manuscript is made available under the Elsevier user license
https://www.elsevier.com/open-access/userlicense/1.0/
way to scale the ammonia synthesis process would be to remove the need for high pressure and

lower the reaction temperature.

There are three major nonconventional processes being researched to activate nitrogen at low

pressure/low temperature during ammonia synthesis: electrochemical, plasma, and

electromagnetic routes. Although a large effort is being placed on the electrochemical approach,

it has issues with low ammonia yields and poor efficiencies that will have to be overcome [4-6].

Non-thermal plasmas have been studied as an alternate reaction route to breaking the N2 bond

into active radicals. While this method is promising, there are difficulties with maintaining

plasmas at atmospheric pressures efficiently. Adding a catalyst to the system is challenging, and

the ammonia yields are low in comparison with other methods [7-9].

Electromagnetic radiation in the form of microwaves has become a focus for chemical

synthesis in recent years due to the increased selectivity and decrease in reaction times and

temperatures [10-12]. Microwaves have the unique ability to selectively heat the catalyst and/or

active sites which allows the reaction to occur at lower bulk temperatures. Microwaves have

also been shown to decrease the activation energy needed for the reaction and hence shift the

equilibrium to more products and/or different selectivity in products. Reactions and heating will

be different between bulk catalyst and dispersed catalyst with dispersed catalyst being able to

take advantage of the selective heating that microwaves offer.

This study tested a commercial Haber-Bosch catalyst that will be considered a bulk iron

system with two supported metal catalysts. Fe and Ru were chosen due to their high activity for

ammonia production. The iron catalyst was developed and tested as a lower cost alternative to

the common Ru-based catalyst and to test a hypothesis of electron movement through the
material systems. This is the first known reporting of ammonia synthesis by a non-plasma

microwave system at atmospheric pressure.

2. Experimental

2.1 Catalyst synthesis

Commercial Haber-Bosch iron based catalyst was obtained from a commercial supplier (not

identified upon request of supplier). Aluminum oxide (Al2O3, 99.9%), ruthenium(III)

nitrosylnitrate (Ru(NO)(NO3)3, Ru 31.3%), and iron (II, III) oxide (Fe3O4, 99.9%) were

purchased from Alfa Aesar and were used without any further purification.

Conventional incipient wetness impregnation was used to prepare Ru catalysts. Typically,

anhydrous Ru(NO)(NO3)3 salt corresponding to 4wt% Ru loading was dissolved in deionized

water and was added dropwise to Al2O3 support. The powder was mixed using a mortar and

pestle for 15 min before drying it at 105˚C overnight. The powder was then calcined in air at

500˚C for 4 hours.

A 20wt% Fe catalyst was produced by solid-state mixing of iron (II, III) oxide with alumina.

The powders were ball milled overnight and calcined in air at 500°C for one hour.

2.2 Catalyst Characterization

Powder X-ray diffraction analysis was performed on a Panalytical X’pert Pro (PW3040) X-

ray diffraction system utilizing Cu Kα radiation. Samples were placed on a zero diffraction Si

holder and were scanned from 5 to 70° (2θ). Analysis was carried with Highscore Plus Analysis

software equipped with a standard ICDD X-ray diffraction database supplied by Panalytical.
 2.3 Microwave reactor

Microwave experiments were performed using a 2.45 GHz (2 kW) magnetron powered

single mode microwave cavity from Sairem (model GMP20K, Neyron Cedex, France). The

schematic of the MW reactor is provided in Figure 1.

The reactions were carried out using the temperature controlled mode with maximum power

set at 300 W. This mode allows for the power to be pulsed to maintain the set temperature. A

quartz tube reactor with a 14 mm internal diameter was placed inside the microwave cavity. The

temperature was measured using a laser-aimed IR pyrometer from Micro-Epsilon (model

CTLM3, Ortenburg, Germany), with a temperature range from 200 to 1500 oC. This pyrometer

setup allowed for the temperature reading from the side of the sample surface. Typically, 2 g of

catalyst was used for each reaction while flowing a 3:1 stoichiometric mixture of hydrogen and

nitrogen with an Ar trace (99.999 %, UHP, Butler Gas) at a fixed flow rate of 100 sccm through

the quartz tube. Reduction of the catalysts were carried out in the microwave at 250 °C for 2hrs

at 300 W for the Ru based catalyst and 500 °C for 2hrs at 300 W for the Fe based catalyst with

the synthesis gas flowing at 100 sccm. Ammonia synthesis tests were conducted from 300 °C

and 500 °C and held at steady state for 30 min at each testing temperature.

2.4 Product analysis

All reactants and products (H2, N2, H2O and NH3) gases were analyzed online using a 200

amu scanning magnetic sector mass spectrometer from Thermo, model Prima BT Benchtop MS

and an Agilent 3000A Micro GC equipped with a TCD detector and four columns: molecular

sieve 5˚A, Plot Q, Plot U, and Alumina. The ammonia production was represented as µmol of
ammonia produced per gram of catalyst per hour and was calculated using the average ppm

recorded by the GC.

3. Results and Discussion

The crystallinity of iron based catalysts, the commercial Harbor Bosh (HB) and supported Fe

catalysts, were studied before and after the reaction. Figure 2 showed the XRD patterns of these

catalysts. The results revealed that the fresh bulk Fe catalyst (HB) is mainly composed of

magnetite (Fe3O4) with no other mixed oxide phase that could be observed. After microwave

irradiation, under a reductive reaction mixture of H2 and N2, almost a complete reduction was

observed. Mainly, metallic Fe with some FeO crystalline peaks were identified, which could be

due to handling the spent catalyst under air. Similarly, the supported Fe catalyst was fully

reduced from Fe3O4 to metallic Fe after microwave irradiation under same reaction gases. The

average iron crystallite size was calculated from Debye Scherrer equation for bulk HB and

supported catalysts and was found to be 31.2 and 34.0 nm for fresh and 65.8 and 30.5 nm for

spent bulk and supported Fe catalysts, respectively. In general, crystallite size did not grow for

the supported Fe catalyst after the reaction, possibly due to difference in heat transfer

mechanisms between supported and bulk iron catalyst. Supported iron particles could possibly

dissipate some of the localized microwave heating into the alumina support at a faster rate

compared to bulk iron particles based on differences in the dielectric properties between the

catalyst and support. These results agree with the observed high reflected microwave power

during ammonia synthesis for supported iron catalyst compared to bulk iron catalyst.
 Figure 3 illustrates the XRD of fresh and spent supported Ru catalyst. The RuO2 crystalline

phase peaks were observed along with broad amorphous peaks corresponding to gamma alumina

support at 2θ = 45.6 and 67.3°. Similarly, a microwave-assisted reduction of this catalyst was

observed, where supported RuO2 particles were reduced into metallic Ru. Interestingly, a partial

phase transformation from gamma to alpha phase of the alumina support was observed after

microwave irradiation. This could be due to the selective heating of acidic polar OH groups in

gamma alumina [13] followed by dehydration reactions and structural reorganization leading to a

new active interface between metallic Ru catalyst and crystalline alpha alumina support. The

presence of crystalline alpha alumina along with metallic Ru particles observed in Figure 3 after

the reaction could be evidence of microwave selective heating at the metal/support interface

since gamma alumina did not convert entirely into alpha phase. These phase changes within

alumina substrate requires high localized temperatures (as alpha phase exist at higher

temperatures compared to gamma phase[14]) which is provided mainly under microwave at the

Ru/alumina interfaces. Boumaza et al. [15] have used XRD and IR spectroscopy to study the

transitions of boehmite (AlO(OH)) to different alumina phases under different temperatures

using conventional thermal treatment. They observed the sequence transition from boehmite to

metastable alumina phases (γ then δ then θ) and finally to highly stable α-Al2O3 phase with

increasing the temperature from 600 to 1300 ℃. This could indicate that during microwave

synthesis of ammonia, hot spots or thermal runaways likely occurred at high temperatures (>

1300 ℃) to convert some of γ to α-Al2O3 as observed in the XRD results. The average

crystallite size was calculated and found to be 27.3 nm for fresh and 21.2 nm for spent Ru

catalysts, respectively. Similarly, the crystallite size of supported Ru particles did not change

significantly after the reaction, like the supported Fe catalyst in Figure 2. Dispersion of Ru or Fe
particles could prevent them from further growth during reaction whereas crystal growth was

observed in case of commercial Fe catalyst.

The average ammonia production rates for the three catalysts are shown in Figure 4. The Ru

based catalyst performed better at lower temperatures compared to the bulk HB catalyst. The

ammonia production of the bulk Fe catalyst was relatively independent of temperature,

remaining around 17µmol/g/h from 300-500°C, and did not produce ammonia under 300°C. The

activity of the supported Fe catalyst was significantly higher compared to the bulk Fe catalyst.

The ammonia production was increased by a factor of 10 at each studied reaction temperature.

Unlike the bulk Fe catalyst that is relatively temperature independent, the supported Fe catalyst

has a significant increase in activity between 300-400°C and then started to decline as the

temperature reaches 500°C. The supported Fe catalyst performed best at 400°C with an average

of 128 µmol/g/h and single pass concentrations up to 950 ppm of ammonia. These reaction

rates, obtained under microwave reactions, is relatively higher than other nonconventional

ammonia synthesis approaches. Peng et al. used a non-thermal plasma created by a dielectric

barrier discharge (DBD) coupled with a Ru based catalyst that achieved a highest single pass

conversion of about 90 ppm at 7 kV, with the measure reaction temperature ranging from 100-

150°C[8]. The supported Fe catalyst presented here has a single pass conversion of 1100 ppm at

400°C without the use of a plasma to dissociate the N2.

The bulk Fe catalyst, consists mainly of magnetite and had a small percentage of promoters

like alumina and calcium. Once reduced, the bulk catalyst can be considered as a network of

iron particles [16] making it behave as a bulk conductive catalyst within the microwave. In a

microwave field, the free electrons in the conduction band of metals will move to the surface,
polarizing the metal at the microwave field’s frequency [17]. The bulk Fe catalyst is essentially

a network of free electrons that move to the catalyst bed’s surface leaving the surface negatively

charged at one end for half of the cycle and the other end for the other half. Those electrons are

then free to react with the nitrogen to aid in the ammonia synthesis as seen in Figure 5.

The supported metal catalysts have dispersed metal particles on a microwave transparent

support. Once placed in the electric field, each individual metal site is polarized leaving them

able to donate electrons to facilitate the nitrogen adsorption process.

This electron movement within the metal is believed to be the reason for poor reactivity of

the bulk Fe catalyst. The bulk Fe based catalyst also has an increased ability to generate heat

through this electron movement compared to the supported catalysts which could be explained

by the low energy input needed during the reaction to maintain the reaction temperature. This

hypothesis was confirmed by dispersing the Fe in alumina disrupting the conductive network in

the case of the 20wt% supported Fe catalyst which not only increased the ammonia production as

compared to the bulk Fe catalyst, but it also doubled the performance of the Ru based catalyst.

This may be due in part to the higher number active sites, 20wt% compared to 4wt% Ru,

however, Ru and Fe do not have the same catalytic activity towards ammonia synthesis.

It is likely that the activity of both supported Fe and Ru catalysts could be improved by the

addition of promotors such as cesium and size and loading optimization of active the metal

sites[18]. These promotors have not only been shown to increase ammonia production due to

increase in acid sites but also increase microwave absorption which would increase the overall

efficiency of the catalyst [19, 20]. It has also been shown that the support material plays an

important role in the synthesis of ammonia for conventional Haber-Bosch processes and the
response of metal oxides vary within a microwave field [21, 22]. Further studies will be needed

to find the best support system for both ammonia synthesis and microwave interactions.

4. Conclusions

It was shown that by using microwave irradiation, ammonia could be produced at low

temperatures (300°C) at ambient pressure using supported metallic catalyst systems. This is the

first known reporting of ammonia being produced at low temperatures and pressures by a non-

plasma microwave system. A 4wt% Ru on alumina catalyst could produce up to 49 µmol/g/h at

300°C while a 20wt% Fe on alumina produced 128 µmol/g/h at 400°C. Microwave reactions

favored the supported catalysts over bulk catalysts due to electron mobilization and selective

heating. Further investigation will be done on optimizing wt% metal loading, particle size, and

promotors used for ammonia synthesis.

Acknowledgements

This project was supported in part by an appointment to the Science Education Programs at

National Energy Technology Laboratory (NETL), administered by ORAU through the U.S.

Department of Energy Oak Ridge Institute for Science and Education. This project was also

supported in part by ARPA-E.

The authors also would like to acknowledge John Hu and John Johnson for their assistance.
References

[1] H. Liu, Ammonia synthesis catalyst 100 years: Practice, enlightenment and challenge, Chinese Journal
of Catalysis, 35 (2014) 1619-1640.
[2] R. Schlögl, Ammonia Synthesis, Handbook of Heterogeneous Catalysis, Wiley-VCH Verlag GmbH &
Co. KGaA2008.
[3] K. Tamaru, Catalytic ammonia synthesis: fundamentals and practice, Plenum Press, New York, 1991.
[4] I. Amar, R. Lan, C. Petit, S. Tao, Solid-state electochemical synthesis of ammonia: a review, J. Solid
State Electrochem, 15 (2011) 1845-1860.
[5] V. Kyriakou, I. Garagounis, E. Vasileiou, A. Vourros, M. Stoukides, Progress in the Electrochemical
Synthesis of Ammonia, Catalysis Today, 286 (2017) 2-13.
[6] X. Guo, Y. Zhu, T. Ma, Lowering reaction temperature: Electrochemical ammonia synthesis by
coupling various electrolytes and catalysts, Journal of Energy Chemistry, 26 (2017) 1107-1116.
[7] T. Mizushima, K. Matsumoto, J. Sugoh, H. Ohkita, N. Kakuta, Tubular membrane-like catalyst for
reactor with dielectric-barrier-discharge plasma and its performance in ammonia synthesis, Applied
Catalysis A: General, 265 (2004) 53-59.
[8] P. Peng, Y. Cheng, R. Hatzenbeller, M. Addy, N. Zhou, C. Schiappacasse, D. Chen, Y. Zhang, E.
Anderson, Y. Liu, P. Chen, R. Ruan, Ru-based multifunctional mesoporous catalyst for low-pressure and
non-thermal plasma synthesis of ammonia, International Journal of Hydrogen Energy, 42 (2017) 19056-
19066.
[9] J. Nakajima, H. Sekiguchi, Synthesis of ammonia using microwave discharge at atmospheric pressure,
Thin Solid Films, 516 (2008) 4446-4451.
[10] E. Altman, G.D. Stefanidis, T. van Gerven, A.I. Stankiewicz, Process Intensification of Reactive
Distillation for the Synthesis of n-Propyl Propionate: The Effects of Microwave Radiation on Molecular
Separation and Esterification Reaction, Industrial & Engineering Chemistry Research, 49 (2010) 10287-
10296.
[11] C.R. Strauss, D.W. Rooney, Accounting for clean, fast and high yielding reactions under microwave
conditions, Green Chemistry, 12 (2010) 1340-1344.
[12] T. Durka, T. Van Gerven, A. Stankiewicz, Microwaves in Heterogeneous Gas-Phase Catalysis:
Experimental and Numerical Approaches, Chemical Engineering & Technology, 32 (2009) 1301-1312.
[13] S. Lamouri, M. Hamidouche, N. Bouaouadja, H. Belhouchet, V. Garnier, G. Fantozzi, J.F. Trelkat,
Control of the γ-alumina to α-alumina phase transformation for an optimized alumina densification,
Boletín de la Sociedad Española de Cerámica y Vidrio, 56 (2017) 47-54.
[14] P.S. Santos, H.S. Santos, S.P. Toledo, Standard transition aluminas. Electron microscopy studies,
Materials Research, 3 (2000) 104-114.
[15] A. Boumaza, L. Favaro, J. Lédion, G. Sattonnay, J.B. Brubach, P. Berthet, A.M. Huntz, P. Roy, R. Tétot,
Transition alumina phases induced by heat treatment of boehmite: An X-ray diffraction and infrared
spectroscopy study, Journal of Solid State Chemistry, 182 (2009) 1171-1176.
[16] M. Appl, Process Steps of Ammonia Production, Ammonia, Wiley-VCH Verlag GmbH2007, pp. 65-
176.
[17] J. Sun, W. Wang, Q. Yue, Review on microwave-matter interaction fundamentals and efficient
microwave-associated heating strategies, Materials, 9 (2016).
[18] T.D. Musho, C. Wildfire, N.M. Houlihan, E.M. Sabolsky, D. Shekhawat, Study of Cu 2 O particle
morphology on microwave field enhancement, Materials Chemistry and Physics, 216 (2018) 278-284.
[19] K.-i. Aika, Role of alkali promoter in ammonia synthesis over ruthenium catalysts—Effect on
reaction mechanism, Catalysis Today, (2016).
[20] Z. Kowalczyk, Effect of potassium on the high pressure kinetics of ammonia synthesis over fused
iron catalyst, Catalysis letters, 37 (1996) 173-179.
[21] E. Reguera, C. Diaz-Aguila, H. Yee-Madeira, On the changes and reactions in metal oxides under
microwave irradiation, J Mater Sci, 40 (2005) 5331.
[22] O. Hinrichsen, F. Rosowski, A. Hornung, M. Muhler, G. Ertl, The kinetics of ammonia synthesis over
Ru-based catalysts: 1. The dissociative chemisorption and associative desorption of N2, Journal of
Catalysis, 165 (1997) 33-44.

Figure Captions

Figure 1: Schematic of flow-through microwave reactor. Powder placed within the fixed bed

region and held in place with quartz wool.

Figure 2: XRD of the fresh and spent bulk HB catalyst and the supported Fe catalyst.

Figure 3: XRD patterns for fresh and spent supported Ru catalyst used during microwave

synthesis of ammonia.

Figure 4: Comparison of average ammonia produced at various temperatures with 300 W power

applied with the bulk iron catalyst and the two supported metal catalysts.

Figure 5: Schematic of electron movements for a)Fe based commercial catalyst and b) 4wt% Ru

based catalyst in an electric field.