Catalytic Reduction of Hydrazine to Ammonia

• Ruvanthi Kularatne

Catalytic Reduction of Hydrazine to Ammonia: The Site of Reduction in Nitrogenase

Abstract

The conversion of N₂ to NH₃ is done mainly via anaerobic bacteria. The enzyme nitrogenase, which can be found in these anaerobic bacteria, is responsible for this conversion. Much research has been conducted in order to identify the structure of the enzyme, the mechanism for the conversion, and the site of reduction. Hydrazine is a substrate and an intermediate of the nitrogenase enzyme. Hence, the reduction of hydrazine to ammonia is used to mimic the late stages of the biological nitrogen fixation. Here the main focus is to identify the metal atom to which the hydrazine molecule binds. In order to identify the binding site of N₂ is Fe, a tris(thiolato)phosphine ligand, P(C₆H₃-3-Me₃Si-2-S)₃³⁻(PS3″), is used as the platform to obtain the iron(II) complex, [P(Ph)₄][Fe(PS3″)(CH₃CN)]. Also, a substrate-bound and product-bound adducts, [N-(Bu)₄][Fe(PS3″)(N₂H₄)] and [N(C₂H₅)₄][Fe(PS3″)(NH₃)] respectively, are synthesized. To determine whether the binding site is the V in vanadium nitrogenase, [P(Ph)₄][V(PS3′′)(Cl)] and [P(Ph)₄][V(PS3′)(Cl)] [PS3′ = P(C₆H₃-5-Me-2-S)₃^3-] are synthesized.

Introduction

Nitrogen is an essential element in all living organisms. It is a major element in nucleotides and in amino acids which ultimately forms DNA and RNA, and proteins respectively. These are the building blocks which make up the nuclei in living organisms. The major source of nitrogen is atmospheric N₂. It is a stable molecule and it has to be converted to a form which can be utilized by organisms. The natural way of nitrogen fixation is by lightening and by anaerobic bacteria, the latter being the most prominent. About 25 % is fixed by the industrial Haber process, which occurs at high temperatures and pressure, whereas the biological processes occur at ambient conditions¹. During the process, N₂ is converted to NH₃, which is a more usable form than N₂. Nitrogen fixation by anaerobic bacteria is catalyzed by the enzyme nitrogenase.

The enzyme is composed of two protein subunits, a MoFe protein and a Fe protein. Studies reveal that the substrate binding and activation in the enzyme occurs at a Mo/Fe/S center. The structure of this molybdenum nitrogenase has been characterized by X-ray crystallography.² The Fe protein has two bound MgATP molecules. During the reduction of N₂, an electron from this Fe protein is transferred to the MoFe protein, which is associated with the hydrolysis of the two MgATP molecules.³ There are reports of three forms of nitrogenase with Mo, Fe and V.⁴ The Fe and the V are also known as the “alternative” forms of nitrogenase¹. The first has a V in place of Mo and the other is an “all-Fe” nitrogenase¹. Although the structures have been identified, the exact mechanism of the catalysis of N₂ by the enzyme is still not fully understood. As a result, research is being conducted to obtain the mechanistic information of nitrogenase. Large number of coordination compounds has been proposed as possible structural or functional models for nitrogenase. Mononuclear and binuclear transition metal complexes and polynuclear Fe/Mo/S aggregates are among the suggested compounds. Hydrazine is a substrate and an intermediate of the nitrogenase enzyme. Hence, the reduction of hydrazine to ammonia is used to mimic the late stages of the biological nitrogen fixation. For the reduction of hydrazine, a proton source and an electron source is necessary (eq 1).¹

N₂H₄ + 2e^– + 2H⁺ ï‚® 2NH₃(1)

Studies through hydrazine have suggested that the site of binding of N₂ is at Fe in the MoFe-cofactor.⁵ However, some research also shows that the reduction site is at Mo in the MoFe-cofactor^1,6 or in a V^II state in vanadium nitrogenase.⁷ Based on electron density maps and X-ray crystallography, it has been found that the Fe/Mo/S cofactor has an elongated MoFe₇S₉ cluster which is composed of MoFe₃S₃ and Fe₄S₃ cuboidal subunits bridged by two or three sulfide ligands.^1,6

In order to identify the site of reduction of nitrogenase and the mechanism involved in the reduction process, much research has been carried out by the formation of various metal complexes. Here, to see if the binding site is Fe, a tris(thiolato)phosphine ligand, P(C₆H₃-3-Me₃Si-2-S)₃³⁻(PS3″), is used as the platform to obtain the iron(II) complex, [P(Ph)₄][Fe(PS3″)(CH₃CN)] (A).⁵ Also, a substrate-bound and product-bound adducts, [N-(Bu)₄][Fe(PS3″)(N₂H₄)] (B) and [N(C₂H₅)₄][Fe(PS3″)(NH₃)] (C), are synthesized. To determine whether the binding site is the V in vanadium nitrogenase, [P(Ph)₄][V(PS3′′)(Cl)] (D) and [P(Ph)₄][V(PS3′)(Cl)] (E) [PS3′ = P(C₆H₃-5-Me-2-S)₃^3-] are synthesized.

Methods

Synthesis of [P(Ph)₄][Fe(PS3″)(CH₃CN)]:

FeCl₂ was added to a solution of H₃[PS3″] and n-BuLi in acetonitrile in the ratio of 1:1:3 respectively, to give an emerald solution. To this [P(Ph)₄]Br in acetonitrile was added followed by ether, and then the solution was placed at −30°C for 3 days. This yielded an emerald crystalline solid of [P(Ph)₄][Fe(PS3″)(CH₃CN)]·4CH₃CN·(C₂H₅)₂O.

Synthesis of [N-(Bu)₄][Fe(PS3″)(N₂H₄)]:

H₃[PS3″], Li and FeCl₂ was reacted in ethanol in the ratio of 1:3:1 respectively, which gave a green solution. It was followed by the addition of excess N₂H₄·H₂O. Then, [N(Bu)₄]Br was added and the reaction mixture was kept at −15°C for 2 days. This resulted in a green crystalline solid of [N-(Bu)₄][Fe(PS3″)(N₂H₄)]·5C₂H₅OH.

Synthesis of [N(C₂H₅)₄][Fe(PS3″)(NH₃)]:

H₃[PS3″], Li and FeCl₂ was reacted in ethanol in the ratio of 1:3:1 respectively, which gave a green solution. Then it was charged with NH₃ gas (1 atm) to generate an emerald solution. Then, [N(C₂H₅)₄]Br was added in ethanol, and the solution was kept at −15 °C for 2 days. A green crystalline solid of [N(C₂H₅)₄][Fe(PS3″)(NH₃)]·3C₂H₅OH was obtained. All the structures were characterized by X-ray crystallography.

Catalytic reactivity of [P(Ph)₄][Fe(PS3″)(CH₃CN)]:

To observe the catalytic activity, an external reductant, [CoCp₂] and a proton source, [LutH][BAr′₄] was used (CoCp₂ = cobaltocene, LutH = 2,6-lutidinium, and Ar’ = 3,5-(CF₃)₂C₆H₃) and all the reactions were carried out in a N₂ enivironment. First, [P(Ph)₄][Fe(PS3″)(CH₃CN)] and CoCp₂ was dissolved in CH₃CN in 1:1 ratio of the complex to the reductant. Then, N₂H₄ and [LutH][BAr′₄] were added to the solution in 1:1:2 ratio (complex: hydrazine: proton source). The reaction was carried out at ambient temperature for about 30 mins. Concentrated HCl was used to quench the reaction. Then, the solvent was removed by vacuum and the solid was extracted with distilled water. Finally, the insoluble residue was removed and the filtrate was taken to do ammonia analysis¹³ and hydrazine analysis.¹⁴

Synthesis of [P(Ph)₄] [V(PS3′′)(Cl)] (D) and [P(Ph)₄] [V(PS3′)(Cl)] (E)

VCl₃(THF)₃ in THF, H₃[PS3″] in methanol and Li were reacted together in a 1:1:3 ratio. This gave a deep red solution. Then, PPh₄Br in CH₂Cl₂ was added and it was layered with pentane. Which gave a red crystalline solid of D. E was synthesized using the same procedure but using the H₃[PS3′] ligand.

Catalytic reactivity of [P(Ph)₄] [V(PS3′′)(Cl)] (D) and [P(Ph)₄] [V(PS3′)(Cl)] (E)

The catalytic reduction of hydrazine by D and E were determined using cobaltocene and 2,6-Lut.HCl, using the same procedure as for A.

Results and Discussion

It was identified from X-ray crystallographic data that the three complexes, A, B, and C were crystallized with solvent molecules. Complex A had four CH₃CN molecules, B had five C₂H₅OH molecules and the complex C had three C₂H₅OH molecules. These solvent molecules filled the voids in these structures by the formation of hydrogen bonds. It was also identified that the three complexes has a five coordinate iron(II) center with a trigonal bipyramidal geometry, which was formed by bonding to the PS3″ ligand and to the nitrogen in each ligand (CH₃CN, N₂H₄ and NH₃ in complexes A, B, and C respectively). Complexes D and E also show a trigonal bipyramidal geometry at the vanadium(III) center in the same manner as in A, B, and C. This can be seen in the ORTEP diagrams shown in (Figure 1).

The results of the catalytic activity of A, for the reduction of hydrazine to ammonia are given by Table 1, those for D are given in Table 2. According to Table 1, the maximum conversion ~83 % is obtained at 30 mins for the catalyst A. For D, ~83 % conversion was obtained after 24 hrs. But a conversion percentage of 90 was obtained after 48 hrs. A controlled reaction was carried out in the absence of complex A. For that reaction, only less than 5 % of hydrazine was converted to ammonia. According to eq 2, hydrazine can decompose into ammonia and nitrogen.

3N₂H₄ ï‚® 4NH₃ + N₂(2)

To interpret the amount of ammonia formed by the decomposition reaction rather than the reduction, the reactions were carried out for both A and D without using the proton and the electron source. The corresponding data for A are given in Table 3. Accordingly, the conversion to ammonia at 30 mins is only 8 % and it was 15.6 % after 1 hr. Therefore it is safe to assume that the majority of ammonia production for A is carried out by the reduction process. There was no production of ammonia for D in the absence of the proton and the electron source.

Figure 1: ORTEP diagrams of (a) A·4CH₃CN·(C₂H₅)₂O, (b) B·5C₂H₅OH, (c) C·3C₂H₅OH, (d) D and (e) E

Table 1: Production of ammonia by A via the catalytic process at different reaction time.

Table 2: Production of ammonia by D via the catalytic process at different reaction time.

Table 3: Production of ammonia for A by the decomposition of hydrazine.

The isolation of the products B and C, the substrate bound and product bound complexes respectively, suggests that the catalytic reduction takes place at single iron site which is supported by the PS3″ ligand. The mechanism for this can be thought as the bound CH₃CN molecule in complex A is replaced by a molecule of hydrazine to give the substrate bound complex B. At this stage, the N-N bond of the bound hydrazine in the iron (II) center is not activated. Therefore, by the addition of a proton source to protonate the hydrazine molecule would allow for the bond breaking of the N-N bond. Hence the first ammonia molecule will be released and a Fe^IVNH₂ intermediate will be formed. Then, Fe^IVNH₂ will be converted to Fe^IINH₃ by another protonation in the presence of an external electron source. Finally, the second ammonia molecule will be released. This reaction pathway can be shown by Scheme 1.

Scheme 1: The reaction pathway for the catalytic reduction process of A

The catalytic reduction of hydrazine by E did not yield any ammonia. This implies that the bound chloride in E is not exchanged with CH₃CN; instead the complex dissolves in it. However this exchange takes place in D, hence the catalytic activity is visible. The reason for the differences in reactivity for these two complexes, D and E, can be accounted by the two ligands, PS3″ and PS3ï‚¢ respectively. In PS3″ ligand, there are more electron donating substituents than in the PS3ï‚¢ ligand. Therefore, the most electron donating ligand, PS3″ ligand, will donate more electrons to V and will facilitate the replacement of the bound chloride with a CH₃CN molecule. Hence, the exchange will not take place in E. Therefore the reduction of hydrazine will not take place.

Conclusion

In summary, it is possible to say that Fe, in MoFe-cofactor, and V, in vanadium nitrogenase, act as the binding site of hydrazine, an intermediate of nitrogen fixation, mimicking the late stages of the nitrogen cycle. Since both the complexes are formed in a tris(thiolato)phosphine ligand platform, the reactivity of the two complexes are comparable. Hence, by comparing the conversion percentages of the two complexes, A and D, with time, it is possible to conclude that the iron complex (A) is far more efficient than the vanadium complex (D).

For further studies, this research can be extended by including Mo in both these complexes and by the formation of cubanes. This would introduce a more complex nature to the complexes and would represent the enzyme more effectively. Moreover, it is possible to compare the efficiency of Mo, by forming complex with Mo on a thiolate platform.

Research Proposal

Title: Proper Identification of the Site of Reduction in Nitrogenase by the Catalytic Reduction of Hydrazine to Ammonia.

Introduction:

The three forms of nitrogenase with Mo, Fe and V,⁴ have been identified. Yet, the exact mechanism and the site of reduction is still not fully understood. Studies through hydrazine have suggested that the binding sites are at Fe in the MoFe-cofactor,⁵ Mo in the MoFe-cofactor^1,6 or in a V^II state in vanadium nitrogenase.⁷ There has been many debates over this topics and much research has been conducted to identify the exact metal atom on which the binding take place. No research has been conducted by including Fe-Mo and V-Fe together. If these two complexes are formed, we might be able to properly identify the site of binding of N₂ in nitrogenase. The enzyme in question is bulky, which is the nature of an enzyme. Hence, to include this bulkiness in the model compounds, we can use cubanes of complex nature. Furthermore, by optimizing these complexes, we may be able to use them in the industry instead of the Haber process.

Goal:

Identify the proper binding site of hydrazine by including both metal atoms in the complex and to use a more complex environment to properly mimic the catalytic activity of the enzyme.

Aim:

1. Synthesis of MoFe- complex and VFe-complex
2. Synthesis of cubanes of the two mentioned complexes

Methodology:

FeCl₂, MoCl₂, H₃[PS3″] and n-BuLi are mixed in 1:1:2:6 ratio in acetonitrile. After 24 hrs, PPh₄Br in acetonitrile will be added to the reaction mixture. Then, the solution will be layered by the addition of ether. Later, the solution can be kept at -30 ï‚°C for about three days. This will result in a complex with Fe and Mo. To check the catalytic activity, the complex: cobaltocene: N₂H₄: [LutH][BAr′₄] in the ratio of 1:2:1:2 respectively, can be used. First, the complex and cobaltocene are dissolved in acetonitrile. Then, N₂H₄ and [LutH][BAr′₄] in acetonitrile are added to the mixture. The reaction is carried out at ambient temperature for 30 mins. Afterwards, conc. HCl is added to quench the reaction and then the solid will be filtered and removed. Finally the filtrate will be taken and ammonia analysis and hydrazine analysis will be carried out using the indophenol method¹³ and PDMAB¹⁴ method respectively.

References:

1. Demadis, K. D.; Malinak, S. M.; Coucouvanis, D. Inorg. Chem. 1996, 35, 4038.
2. Einsle, O.; Tezcan, F. A.; Andrade, S. L. A.; Schmid, B.; Yoshida, M.; Howard, J. B.; Rees, D. C. Science 2002, 297, 1696.
3. Danyal, K.; Inglet, B. S.; Vincent, K. A.; Barney, B. M.; Hoffman, B. M.; Armstrong, F. A.; Dean, D. R.; Seefeldt, L. C. J. Am. Chem. Soc. 2010, 132, 13197.
4. Malinak, S. M.; Demadis, K. D.; Coucouvani, D. J. Am. Chem. Soc. 1995, 117, 3126.
5. Chang, Y-H.; Chan, P-M.; Tsai, Y-F.; Lee, G-H.; Hsu, H-F. Inorg. Chem. 2014, 53, 664.
6. Coucouvanis, D.; Mosier, P. E.; Demadis, K. D.; Patton, S.; Malinak, S. M.; Kim, C. G.; Tyson, M. A. J. Am. Chem. Soc. 1993, 115, 12193.
7. Chu, W-C.; Wu, C-C.; Hsu, H-F. Inorg. Chem. 2006, 45, 3164.
8. Demadis, K. D.; Coucouvanis, D. Inorg. Chem. 1995, 34, 436.
9. Demadis, K. D.; Coucouvanis, D. Inorg. Chem. 1995, 34, 3658.
10. Palermo, R. E.; Singh, R.; Bashkin, J. K.; Holm, R. H. J. Am. Chem.Soc. 1984, 106, 2600.
11. Zhang, Y.-P.; Bashkin, J. K.; Holm, R. H. Inorg. Chem. 1987, 26, 694.
12. Wong, G. B.; Bobrik, M. A.; Holm, R. H. Inorg. Chem. 1978, 17, 578.
13. Chaney, A. L.; Marbach, E. P., Clin. Chem. (Winston-Salem, N. C.) 1962, 8, 130.
14. Haji Shabani, A. M.; Dadfarnia, S.; Dehghan, K., Bull. Korean Chem. Soc. 2004,

25, 213.

1