Environ Chem Lett (2012) 10:295–300 DOI 10.1007/s10311-012-0366-2

ORIGINAL PAPER

Urea formation from carbon dioxide and ammonia at atmospheric pressure Xiaofeng Xiang • Li Guo • Xing Wu Xiaoxun Ma • Yashen Xia

•

Received: 17 April 2012 / Accepted: 23 April 2012 / Published online: 22 May 2012 Ó Springer-Verlag 2012

Abstract Urea synthesis, currently the largest use of carbon dioxide in organic synthesis, is conventionally operated at high pressure and high temperature. Here, we report for the first time that urea forms at atmosphere and ambient temperatures by negative corona discharge in gas phase. The conversion of CO2 and yields of a solid mixture of urea and ammonium carbamate, which was identified by the 13C NMR spectrum, rise with reducing temperatures and increasing molar ratios of NH3/CO2 and discharge frequencies. The conversion of carbon dioxide was found to be 82.16 % at 20 °C and 1 atm with a molar flow ratio of n(NH3)/n(CO2) of 2.5. High pressure and high temperature as energy inputs are not necessary. Keywords Electronegative ions  Carbon dioxide  Ammonia  Urea synthesis  CO2 utilization  Anion reaction

Introduction Finding feasible technologies for the utilization of CO2 at the production source, e.g., flue gas which is normally at atmosphere and moderate temperatures, are crucial for

X. Xiang  L. Guo  X. Wu  X. Ma (&) School of Chemical Engineering, Chemical Engineering Research Center of the Ministry of Education for Advanced Use Technology of Shanbei Energy, Shaanxi Research Center of Engineering Technology for Clean Coal Conversion, Northwest University, Xi’an 710069, Shaanxi, People’s Republic of China e-mail: [email protected] Y. Xia (&) HyChar Energy, LLC, Newton, MA 02466, USA e-mail: [email protected]

large-scale reduction of the emission of CO2 (O’Neill et al. 2010; Halmann and Steinberg 1999). Urea synthesis is currently the largest use of carbon dioxide in organic synthesis. Urea, CO (NH2)2, is the most widely-produced and commonly-traded nitrogen fertilizer. Urea is produced industrially by synthesis of ammonia/carbon dioxide technology, which is a two-step process where the ammonia and carbon dioxide react to form ammonium carbamate, which is then dehydrated to urea, operating at around 180–210 °C and nearly 150 atm pressure. This synthesis is generally described as follows, 2NH3 þ CO2 , NH2 COONH4

ð1Þ

NH2 COONH4 ! COðNH2 Þ2 þ H2 O

ð2Þ

The change of the standard Gibbs free energy at 25 °C and 1 atm, for reactions (1) and (2), are -23.8 kJ and ?16.7 kJ, respectively, acting as reversible or kinetically slow processes, with low conversions for the former or being infeasible thermodynamically for the latter at ambience, unless proper external energy input is provided. It is not energy-effective and cost-effective to capture CO2 from production sources, e.g., coal-fired power plants, for urea synthesis, due to the potential high capital costs of the CO2 capture infrastructure. Although it has been suggested (Barzagli et al. 2011) that carbon dioxide could first be captured by ammonia in an organic solvent, e.g., anhydrous ethanol, at 0 °C and 1 atm to obtain ammonium carbamate, before the solid product is used to produce urea, the major concerns remain on availability and the costs of huge amounts of organic solvent in the production sources. While it has been considered that a method of non-equilibrium or non-thermal plasma-aided treatment of CO2 could be used for CO2 decomposition and methane reforming (Pietruszka and Heintze 2004; Liu et al. 1999;

123

296

Environ Chem Lett (2012) 10:295–300

Christophorou 1984), thus realistically expecting to avoid elevated temperatures and/or pressure and the thermodynamic restrictions of equilibrium compositions by this technique, high temperatures and catalysts would still be needed, because the reforming of methane would require prohibitively high amounts of energy (12.6 eV for CH4) for ionization. We intend to show another CO2 conversion route as a more efficient option for chemical synthesis, i.e., applying electronegative ions (Stoffels et al. 2002; Zhukhovitskii et al. 2003), by which we can use an electron-discharge technique, e.g., negative corona discharge (Chen and Davidson 2003) to attach electrons to polar gaseous molecules. These do not normally ionize to both positive and negative ions under low energy corona discharge, so creating stable unipolar or gaseous anions as reducing agents for the reduction of CO2, avoiding side reactions and restriction of thermodynamic equilibrium on conversion. Here, we report how urea and ammonium carbamate may be synthesized by using negative corona discharge. We applied a commercially available negative corona generator (SJ-2000E, 3 kW), which is normally used for surface modification or treatment of non-paper substrates (e.g., plastics and foils), with an output voltage of -15 kV and current of 0–2 A, to discharge electrons into a glass-tube reactor, which is 25 mm ID and 400 mm in length, as shown in Fig. 1. This reactor is filled with PTFE-coated magnetic bars on the upper section, to establish a magnetic field to hold electronegative anions around the surface of the PTEF magnetic bars, so that interactions between the molecules and anions can be enhanced.

T

Heating coil Glass tube Magnetic bars Charge pin

Distributor

NH3

CO2

Fig. 1 Solidified particles on the glass-tube reactor (25 mm ID) bottom after reaction. Inset an image of solid product accumulated on the surface of the metals and glass inside the tube reactor

123

Experimental section Apparatus An SJ-2000E negative power supply with voltage range of -15 kV and power of 3 kW were applied to the experiments. A half-wave rectifier unit was connected to the outlet of the power supply, so as to generate a stable pulsed DC negative high voltage. An S49-33/MT mass-flow meter, and D08-1D/ZM flow display in the range of 0.0–0.5 L/min, was used. A thermostar GSD320 mass spectrometer was connected to the outlet of the reactor. A Langmuir probe was made in this laboratory and calibrated by an AIC1000 negative ion concentration detector for measurement of the negative ion concentration. An SXKV digital temperature control of the electric-powered heater and a TGA-IR (Bruker Optics, Germany) infrared spectrum analyzer were also utilized for the experiments. Experimental procedure The reaction device is shown in Fig. 1. In the experiments, the reactor was purged by N2 for 15 min. CO2 (99.99 %) and NH3 (99.99 %) gases were fed from the gas cylinder, through a mass flowmeter, and drying and preheating zones, entering the tubular reactor from the bottom. The gases flow through the lower porous distributor before entering a discharge zone with a length of 50 mm to contact the corona discharge to form mixtures of anions and gases. These mixtures were then pushed through the upper porous distributor as a support to the magnetic bars into the packing bed of magnetic bars, which is 245 mm in length, for anion reaction before leaving the system from the top. In the discharge zone, a bunch of discharge pins, which are made of 20 silver-coated stainless steel needles, of 10 mm ID and 35 mm in length, are inserted from the side of the tube and connected to the external negative high-voltage corona generator. The gaseous and anion components were then analyzed by the Thermostar GSD320 mass spectrometer within the flow rate range of 0.06–0.22 L/min. Our experiment shows that, under corona discharge, when CO2 gas was mixed with ammonia in the reactor, visible white crystal products could be observed in the glass reactor and the outlet, as shown in the inset of Fig. 1. In order to identify the compositions of the solid product, we obtained an SEM image of the crystal particles (Fig. 2a). We used a 13C NMR spectrometer by INOVA400 of Varian, confirming the existence of urea and ammonium carbamate. The proportion of urea in the solid mixture is about 24 % (Fig. 2b). It is evident that the anion reactions and the reaction (1) could co-exist, leading to the formation of solid mixtures of urea and ammonium carbamate. We applied the conventional NaOH titration (Van

Environ Chem Lett (2012) 10:295–300

297

Negative ion concentration , 1012 icon/cm3

(A) 3x10

3

NH 3 2.5x10

2x10

1x10

3

2

n(CO2 ) /n(NH3 )=2.5

2

CO2 0 8.5

9.5

10.0

10.5

11.0

Discharge frequencies ,kHz

Ammonium carbamate

(B)

9.0

NH COONH 2 4 161.463

Fig. 3 Effect of discharge frequencies on the concentration of electronegative ions

Slyke and Cullen 1914) to examine the solid mixture obtained at 20 °C, and found that the proportions of urea in the solid were 12.34, 16.99, and 17.78 wt% for three solid samples collected from different spots inside the glass tube, similar to the results obtained from the 13C NMR spectrum.

meter, and the mass spectrum of reacting gases by online Thermostar GSD320 mass spectrometer, in order to examine the components in the reactor (Fig. 3). Combining the anion density measurement with the mass spectral analysis with the feed of CO2 and NH3 into the reactor, we could deduce the existence of anions of NH3-, NH2-, and O2-. We have found that, with increases of corona discharge frequencies, the concentrations of electron negative ions rise for both ammonia-only (mainly NH3- and NH2-) and mixtures of CO2 and ammonia (mainly NH3-, NH2-, CO2-, and O2-), indicating that, with an increase of discharge, more electrons could attach to ammonia molecules and radicals to form anions, while for CO2-only (mainly CO2-, O2-), the concentrations of the electronegative ions keep really low compared to those of ammonia-only, implying that the anions of CO2-, O2- could only have small contributions for urea synthesis. In order to study the effects of operating conditions on the CO2 and NH3 conversion and molar yield, we carried out some measurements to determine the dependence of conversions on electron discharge, temperatures, the ratios of molar flows, total flow rates, reaction times, and magnetic intensity.

Results and discussions

Effects of discharge frequencies on the CO2 conversions

Urea C=O 163.656 Solvent: CD3OD

180

170

160

ppm 150

Fig. 2 Identification of compositions of solid product from the electronegative anion reactor. a Images of solid products by SEM. b 13C NMR spectrum of solid product shows the characteristic structure of the solid mixture, showing the product as both urea and ammonium carbamate with 24 % of urea

Effects of discharge frequencies on the concentration of electronegative ions In order to examine the gas-phase compositions and changes of electronegative ions with corona discharge, we also measured the anion density in the reactor at 5 cm from the discharge pins by a Langmuir probe and AIC1000 ion

We found that, at no discharge, as a blank test, the CO2 conversion is 9.51 %, indicating that without electron discharge, some ammonium carbamate could also be formed by the reaction (1) at the same time. We examined the dependence of conversions on discharge frequency, finding (Fig. 4a) that, with increases of corona discharge frequencies, the conversion of CO2 and NH3 appears to rise at 20 °C.

123

298

Environ Chem Lett (2012) 10:295–300

With electron discharge, the conversions and molar yields rise at 20 °C, probably because more ammonia anions and radicals and electron attachment may be formed. This means that the yields of the urea is dependent on that of the anions, also implying that enhanced negative ionization might drive a high yield of urea.

(A) 80

100

CO 2

60

NH3

60

NH2 COONH4 (Ammonia carbamate)

40

40

20

NH2 CONH2 (Urea)

20

Yield,%

Conversion,%

80

CO 0

0 8.5

9.0

9.5

10.0

Discharge frequency,kHz

(B) 60

100 90

50

CO2

NH 2 COONH4

NH3

(Ammonia carbamate)

40

70 30

60 50

NH2 CONH2 (Urea)

Yield,%

Conversion %

80

20

40 10

CO

30

0

20 0

10

20

30

40

50

60

70

Effects of N(I2)/N(CO2) on the CO2 conversions

(C) 60

100

NH3

40

NH2 COONH4 (Ammonia carbamate)

60

30

CO2

Yield,%

Conversion, %

50

80

70

20

CO

50

10

NH 2 CONH2 (Urea)

40 0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

n( NH 3 ) /n(CO 2 )

Fig. 4 Dependence of the rate of CO2 conversion and yields of solid products on different conditions. a Effect of discharge frequencies on the CO2 conversion and yields of solid products at 20 °C and n(NH3)/ n(CO2) = 2.5. b Effect of temperatures on the CO2 conversion and yield of the solid products at nNH3/nCO2 = 2.5 and discharge frequency = 10.245 kHz. c Effect of ratio of molar flow rates of N(I2)/N(CO2) on the CO2 conversion and yields of solid mixture at 20 °C, and discharge frequency of 10.245 kHz

123

As for the effect of temperatures on conversions, it is revealed (Fig. 4b) that CO2 and NH3 conversion decreases with increasing temperatures, indicating that this anionassisted reaction process should be exothermic. The lower the temperatures, the greater the yields of urea and ammonium carbamate. Additionally, after 60 °C, the molar yields of urea and ammonium carbamate decrease quickly, probably due to decomposition of the ammonium carbamate into CO2 and NH3. We also noticed that the yield of CO decreases dramatically after the ratio is greater than 2.5, probably due to consumption by the formation of more solid. We found that the conversions of CO2 could reach 82.16 % and the molar yield of the whole solid mixture of both urea and ammonium carbamate is 54 % at 20 °C and 1 atm, at the gas flow rate of 0.12 L/min, the negative corona discharge frequency of 10.245 kHz, and n(NH3)/ n(CO2) of 2.5.

80

Temperature,

90

Effects of temperature on the CO2 conversions

We also evaluate the dependence of conversions on the ratios of molar flow rates, finding (Fig. 4c) that with the increase of the ratios from 1 to 3.5, conversions of CO2 and molar yields of urea and ammonium carbamate increase, while those of NH3 keep decreasing and CO rises before the ratio is 2.0 and decreases after it is greater than 2.0, implying that CO consumed by the reaction to form more solid products of urea and ammonium carbamate. According to the stoichiometric ratio of 2, the urea synthesis may require at least 2 mol of NH3 to form 1 mol ammonium carbamate or urea. From the results, it appears consistent to the Le Chatelier principle for a thermodynamically equilibrium system that slightly higher ratios can be beneficial for the higher conversions of CO2, although the present reaction system is not in full equilibrium due to the existence of active anions and electrons, implying that, for a dilute anion reaction, the Le Chatelier principle could stand. Remarks in reaction mechanism It has been noted (Rienstra-Kiracofe et al. 2002; Nielsen and Bradbur 1937; Spencer and Gallimore 2010) that, although ammonia does not have an electron affinity because of its closed electronic shells, it has strong

Environ Chem Lett (2012) 10:295–300

299

attractive polarization interactions with an excess electron, then binding the excess electron and releasing energy. It is expected (Gutsev et al. 1998) that, under low-energy negative corona discharge, NH3 could form anions, probably NH3-, NH2-, and CO2 as typical electronegative gases (Chen and Davidson 2003; Gutsev et al. 1998), and could obtain an excess electron to form CO2– and also partially decompose into CO and oxygen anions, O2–, in low energy discharge. It is known that the work needed to remove electrons from the corona electrode surface is approximately 4 to 5 eV (Chen and Davidson 2003) for the metals most likely to be used in corona discharge devices. We speculate that, once the anions and radicals formed under electron discharge meet, the following process may occur (Sommerfeld et al. 2004):  2NH 3 þ CO2 ! NH2 COONH4 þ 2e

2NH3 þ

CO 2

! NH2 COONH4 þ 2e



 2NH 3 þ CO ! COðNH2 Þ2ðsÞ þ H2 þ 2e

1  2NH2 þ CO 2 ! COðNH2 Þ2ðsÞ þ O2 þ 2e 2 1  2NH 2 þ CO2 ! COðNH2 Þ2ðsÞ þ O2 þ 2e 2  2NH 2 þ CO ! COðNH2 Þ2ðsÞ þ 2e

ð3Þ ð4Þ ð5Þ ð6Þ ð7Þ

discharge, electronegative ammonia anions and radicals can reduce CO2 into solid urea, without any metal catalyst, and that the conversion is found to be up to 82.16 % at 20 °C and 1 atm. We also observed that the conversion of CO2 and yields of the solid mixture of urea and ammonium carbamate increases with reducing temperatures, and with increasing molar ratios of NH3/CO2 and discharge frequencies. It is now clear that high pressure and high temperature as energy inputs are not necessary for urea synthesis. Furthermore, we believe that the electron-attaching method should be suitable for other electronegative gases in anion-assisted treatment of CO2 at ambience so as to implement CO2 utilization. More generally, we also infer that other electronattachable gas-phase syntheses or reactions that conventionally need high pressure and high temperature as input energy may also be implemented by this approach. Acknowledgments This work was financed by Shaanxi Important Innovative Projects in Science & Technology of China (2009ZKC0406) & (2010ZDKG-43), the National Key Technology R&D Program of China (2009BAA20B02), the Key Science and Technology Program of Shaanxi Province of China (2010K01-082), Project supported by the National Natural Science Foundation of China (NSFC21006078), and Interdisciplinary funded projects of Northwest University Postgraduates (09YZZ52).

ð8Þ

We believe that the proportion of the solid urea, at about 1/4 in the whole solid mixture in the present work, can be attributed to the limited dissociation of NH3 and CO2, also implying that any method for enhancing dissociation of NH3 for forming amide anions and radicals, which would be needed by reactions (5)–(8) for urea synthesis, should potentially help the yield of urea. We also infer that there could exist a principle for such anion–gas reactions that, as long as the input energy by electron attachment can be larger than the Gibbs free energy differences for those reactions, the process could potentially be driven forward to products with significant conversions. Also, it could potentially be more effective in combination with proper catalysts (Pietruszka and Heintze 2004), which could be excited by plasma discharge, for more electric current and better selectivity of the desired products.

Conclusions In the present work, we have shown for the first time that, although reduction of CO2 by NH3 into urea at ambience would not be feasible by conventional processes, once two gases are under the negative corona discharge, urea can form at ambience. We found that, under negative corona

References Barzagli F, Mani F, Peruzzini M (2011) From greenhouse gas to feedstock: formation of ammonium carbamate from CO2 and NH3 in organic solvents and its catalytic conversion into urea under mild conditions. Green Chem 13(5):1267–1274 Chen J, Davidson JH (2003) Model of the negative DC corona plasma: comparison to the positive DC corona plasma. Plasma Chem Plasma Process 23(1):83–102 Christophorou LG (1984) Electron–molecule interactions and their applications. Academic, New York Gutsev GL, Bartlett RJ, Compton RN (1998) Electron affinities of CO2, OCS, and CS2. Chem Phys 108:6756–6763 Halmann MM, Steinberg M (1999) Greenhouse gas carbon dioxide mitigation. CRC, Boca Raton Liu CJ, Mallinson R, Lobban L (1999) Comparative investigations on plasma catalytic methane conversion to higher hydrocarbons over zeolites. Appl Catal A Gen 178(1):17–27 Nielsen RA, Bradbur NE (1937) Electron and negative ion mobilities in oxygen, air, nitrous oxide and ammonia. Phys Rev 51(2):69–75 O’Neill BC, Dalton M, Fuchs R et al (2010) Global demographic trends and future carbon emissions. Proc Natl Acad Sci USA 107(41):17521–17526 Pietruszka B, Heintze M (2004) Methaneconversion at low temperature: the combined application of catalysis and non-equilibrium plasma. Catal Today 90(1–2):151–158 Rienstra-Kiracofe JC, Tschumper GS, Schaefer HF III et al (2002) Atomic and molecular electron affinities: photoelectron experiments and theoretical computations. Chem Rev 102:231–282 Sommerfeld T, Meyer H-D, Cederbaum LS (2004) Potential energy surface of the CO2- anion. Phys Chem Chem Phys 6:42–45

123

300 Spencer LF, Gallimore AD (2010) Efficiency of CO2 dissociation in a radio-frequency discharge. Plasma Chem Plasma Process 31(1): 79–89 Stoffels E, Flikweert AJ, Stoffels WW et al (2002) Plasma needle: a non-destructive atmospheric plasma source for fine surface treatment of (bio)materials. Plasma Sources Sci Technol 11(4): 304–312

123

Environ Chem Lett (2012) 10:295–300 Van Slyke DD, Cullen GE (1914) A permanent preparation of urease, and its use in the determination of urea. J Biol Chem xix:211 Zhukhovitskii DI, Schmidt WF, Illenberger E (2003) Stability of negative ions near the surface of a solid. J Exp Theor Phys 97(3):606–614