Towards nitrogen fixation and
functionalization of dinitrogen beyond
ammonia
Max van Druenen
Supervisor: Andrew Jupp
First assessor: dhr. dr. J.C. Slootweg
2
Summary
Nitrogen fixation and functionalization is a difficult and important challenge in organometallic chemistry. Over the past century, the energy intensive Haber-Bosch process dominated global nitrogen fixation. Due to the rise in awareness of environmental issues, the demand has grown for an environmentally friendly alternative. The strong triple bond of the dinitrogen molecule poses a thermodynamic challenge for promoting reactivity. While most research is focusing on the formation of ammonia, some progress has been made towards generating more complicated nitrogen-containing organic molecules. This review focuses on the recent efforts to functionalize dinitrogen via complete N≡N bond cleavage to form products other than ammonia. Several strategies for the synthesis of nitriles, hydrazines, silylamines, isocyanates and some other nitrogen containing products are summarized and discussed.
Contents
Summary ... 2 Introduction ... 3 Haber-Bosch process ... 4 N2 coordination ... 5 Hydrazine ... 6 Nitriles ... 10 Silylamines ... 15Heterocycles via N(SiMe3)3 ... 17
Isocyanates ... 18
Oxamide ... 23
Diazonium salts ... 24
C-H activation ... 25
Nitrogen fixation at boron ... 26
Lithium ... 27
Discussion ... 29
Conclusions and outlook ... 30
3
Introduction
Nitrogen is an essential component of DNA, proteins and enzymes and is therefore a key component to life on earth. Atmospheric dinitrogen (N2), the most abundant form of nitrogen, is not a suitable nutrient for most organisms. Therefore, conversion of atmospheric nitrogen into reactive nitrogen (Nr) supports biomass growth. Reactive nitrogen compounds include oxidized nitrogen species, reduced nitrogen species, organic bound nitrogen, amine derivatives, as well as organic nitrogen in proteins and other biological substances.1 There are a few main methods to generate reactive nitrogen by nature, the first one being by lightning (globally approx. 5 Tg N yr-1) and biological N2-fixation (globally approx. 198 Tg N yr-1).2 The invention of the Haber-Bosch process drastically increased global nitrogen fixation, which led to a great increase in human population growth.3
The Haber-Bosch (H-B) process revolutionized our production of food after its invention in the early 1900s. By generating ammonia from nitrogen and hydrogen, this precursor to nearly all nitrogen-based compounds could be manufactured efficiently. Since then, human production of reactive nitrogen has had substantial impact on the ecological environment; inefficient use of Nr has led to accumulation in the environment.4,5 This accumulation leads to stratospheric ozone depletion, acid rain, water eutrophication, biodiversity loss, and impairment on human health.5 Currently, the biochemical flow of nitrogen to the biosphere and oceans is one of the most high risk categories of the Planetary Boundaries framework.6 Therefore, redefinition of our Nr use is of great importance. However, ineffective use is not the only environmental issue in Nr production. Due to the H-B process and the steam methane reforming (SMR) for the hydrogen production both having exceedingly high energy costs, a reliance on fossil fuels and a resulting CO2 emission, there have been considerable efforts into finding more environmentally friendly and sustainable alternatives.7
Although most of the pollution could be negated by finding alternatives for the SMR reaction, no commercially viable solution has been found yet. There are a few methods for the generation of hydrogen (e.g. electrochemical water splitting), but these methods are too energy intensive compared to the amount of hydrogen produced. Therefore, the scope should be broadened to finding alternative methods for producing reactive nitrogen from renewable resources in an environmentally friendly manner. The main challenge for these alternative routes is the activation of the strong N≡N bond.7 Research has shown different promising alternatives for the production of ammonia using biochemistry, heterogeneous catalysis, plasma-driven transformations, and electro- and photocatalysis.7 Although these processes have been studied extensively, an industrially applicable alternative to the H-B process is yet to be found.
Approximately 20% of produced ammonia is applied in organic synthesis.8 Redefining this sector of Nr synthesis will have a significant impact on the demand for ammonia. Attempts for finding an alternative to the H-B process mostly focus on the production of ammonia. However, over the past few years, new attempts towards the synthesis of Nr other than ammonia have risen.
Conversion of N2, resulting in other Nr than ammonia, is both fundamentally and practically challenging due to its chemical stability under ordinary conditions.9 The nonpolar N≡N triple bond has a high bond energy of 941 kJ.mol-1, rendering cleavage and hydrogenation extremely difficult.10 In recent years, some work has been published converting dinitrogen to hydrazine,9,11,12 nitriles,13,14 tris(trimethylsilyl)amine,15–17 and even some complexes with cyanate moieties.18,19 This shows that even the high energy barrier of N2 can be overcome and depicts the potential of divergent synthesis routes. These approaches are still in their infancy, but are fundamental in finding alternatives for the H-B process and will continue to be interesting topics in nitrogen chemistry.9,20
4
Haber-Bosch process
The H-B process was invented by Fritz Haber, who sold it to BASF where it was optimized by Carl Bosch from 1909 to 1913. Nowadays, the process is mostly used for the production of fertilizer. Haber and Bosch were both awarded Nobel prizes for their endeavors in 1918 and 1931, respectively.10 Nowadays, most plants use almost identical conditions as the original concept, albeit with slightly lower energy consumption due to catalyst optimization.21
The H-B process normally proceeds under elevated temperature (500 oC) and pressure (>100 bar) in the presence of an Fe-based catalyst to react gaseous nitrogen and hydrogen to ammonia, but the largest amount of energy is consumed for the hydrogen production in the endothermic steam-methane reforming (SMR) at 800 oC-1000 oC, and the reactant purification and compression.4 The extreme reaction conditions of both these reactions lead to a combined energy consumption of 1-2% of the total energy production worldwide and 3-5% of the world’s natural gas.4,22
The H-B process uses N2 and H2 as feedstock. The latter is acquired via the steam methane reforming process. The first step of this process is the production of syngas using a nickel oxide alumina catalyst to obtain CO and H2 (1). However, the methane gas only partially reacts. Oxygen and nitrogen gas are added to increase syngas yield in a secondary reformer (2). The nitrogen is added for the subsequent ammonia synthesis. Finally, the carbon monoxide of the syngas/nitrogen mixture is oxidized to CO2 by the water-gas shift reaction (3), facilitating easy removal from the gas mixture by gas scrubbing with triethanolamine, leaving the desired N2 and H2 mixture for the H-B process.23
CH4 + H2O CO + 3H2 ΔH298 = 206.3 kJ∙mol-1 (1) 2 CH4 + O2 2CO + 4H2 ΔH298 = -71 kJ∙mol-1 (2) CO + H2O CO2 + H2 ΔH298 = -41 kJ∙mol-1 (3)
The N2 and H2 mixture resulting from the steam methane reforming is preheated and transferred to the reactor. Here the gaseous mixture is adsorbed by the catalyst, where the nitrogen forms NH, NH2, and then NH3. The most optimal catalysts for the H-B process are Fe(111) and Fe(211) catalysts, where the surface layer arrangement has highest reactivity due to exposed C7 sites (iron with seven nearest neighbours.23 The rate-determining step is, of course, the nitrogen dissociation due to the nitrogen triple bond being the strongest bond that needs to be broken.
5
N
2coordination
N2 cleavage is a challenging task due to the triple N≡N bond being one of the most stable bonds known. It is the second strongest bond in any diatomic molecule, surpassed only by CO.25 Although N2 is a typically inert diatomic molecule, it exhibits a large and varied coordination chemistry with almost every transition metal in the periodic table.26 There are multiple dinitrogen coordination modes for metal mediated nitrogen cleavage, the most common are: 1) end-on; 2) dinuclear end-on; 3) dinuclear side-on; 4) dinuclear side-on end-on (figure 2).27 Monometallic side-on complexes are metastable species and are therefore extremely rare.28
Figure 2: common coordination modes for dinitrogen27
When bimetallic complexes are formed, the most common coordination mode is the dinuclear end-on coordination. This lead to the theory that this was the preferred arrangement. However, as ligand design advanced further, the dinuclear side-on coordination became more prevalent.29 The coordination modes are determined by the type of metal, its oxidation state, and ligand properties.27 The degree of activation mostly depends on the reduction potential of the metal. Early transition metals with low oxidation states show strong dinitrogen activation, whereas late transition metals generally show weak activation.30 Third-row transition metals have been observed to be exceptions in this activation.31,32
When a monometallic end-on N2 complex is formed, the terminal nitrogen will become negatively charged and thus susceptible for electrophilic attack, whereas the coordinated nitrogen will become more positive and therefore more susceptible for nucleophilic attack.33 In bimetallic complexes, nitrogen becomes nucleophilic due to the strong reduction and charge delocalization over the M2N2 unit.34
In 1995, Laplaza and Cummins showed pioneering work where a three-coordinate molybdenum complex, isolobal with dinitrogen, cleaves the N≡N bond at ambient temperature and pressure to form the corresponding molybdenum nitride.35 The π10 en π8σ2 electron configurations are optimal for N2 cleavage to nitrides via end-on bridging and side-on bridging with bimetallic N2 complexes, according to molecular orbital analyses and symmetry considerations.27 Still, these findings ignore other important factors such as M≡N bond strength, steric effects, high/low spin transitions and π-donating ligands.34 These factors all contribute to more advanced predictions for possibilities of N2 cleavage. Since then, a large number of metal complexes have been identified that promote N2 cleavage under both thermal and photochemical conditions.26
6
Hydrazine
Hydrazine is an interesting target, due to its high reactivity and facile catalytic reduction to ammonia.36,37 Although hydrazine synthesis from molecular nitrogen has been reported as early as in 1969, the amount of research is scarce.11,38 Hydrazine is currently synthesized from ammonia and hydrogen peroxide in the so-called peroxide process. In this process ammonia is oxidized by hydrogen peroxide in the presence of methylethylketone (MEK) and catalytic acetamide.39
Initially, MEK forms a Schiff base with ammonia, and acetamide forms ethanimidoperoxoic acid with the hydrogen peroxide. The ethanimidoperoxoic acid and the Schiff’s base form oxaziridine and acetamide, which is thus regenerated. The oxaziridine condenses with a second molecule of ammonia to give the hydrazine, which condenses with another equivalent of MEK to give methyl ethyl ketazine in high yield. The mixture then undergoes phase separation and the organic layer is distilled, followed by hydrolysis to obtain hydrazine hydrate. The aqueous layer is dehydrated to obtain the MEK and the expensive acetamide.39
Figure 3: General procedure of the peroxide process for the generation of hydrazine39
The high yielding and ecofriendly peroxide process diminished the urgency for a different approach. Nevertheless, nitrogen fixation was already a topic of interest in the sixties.38,40 In this period, a few reports were made, focusing on enzymatic nitrogen fixation (the process involving complete or partial dissociation of the N≡N bond), rather than obtaining hydrazine in high yields.40 These processes mostly produced ammonia and yielded hydrazine as a byproduct.
There are only a few examples of direct synthesis of C-N bonds of hydrazine derivatives using dinitrogen as nitrogen source. In 2007, Chirik and co-workers reported a bihafnocene complex mediated synthesis of substituted hydrazine from dinitrogen and carbon dioxide.41 Their [(η5 -C5Me4H)2Hf]2(µ2, η2, η2-N2) (4) complex yielded a product with a new carbon-nitrogen bond upon addition of 2 equivalents of CO2 under inert conditions at RT in toluene (5) with a 41% yield (figure 4). The substituted hydrazine was removed from the complex by silylation with a surplus of Me3SiI to obtain (SiMe3)2NN(CO2SiMe3)2 in 85% yield. The resulting diiodo hafnium complex 7 could be transformed into complex 4 by reductive nitrogen fixation using sodium amalgam with a yield of 27%.42
7
Figure 4: Substituted hydrazine synthesis by Chirik and co-workers41
This reaction pathway was later adopted in the same institute using zirconium for the metal complex.12 Similar results were obtained initially, but when a C2-symmetric ansa-zirconocene complex was employed, selective insertion of 2 equivalents of CO2 was promoted. This resulted in the synthesis of a variety of N,N’-dicarboxylated hydrazines from two inert gases (CO2 and N2).
The highest yielding report of this method is by Kawaguchi and co-workers, who managed to obtain a quantitative yield in the carboxylation step using a titanium complex at RT.43 The reaction time was optimized to 0.5 h, showing great improvement over the method by Chirik and co-workers, which had a reaction time of 3-5 days. Silylation of the carboxylated complex using chlorotrimethylsilane allowed the isolation of a mixture of carboxyl hydrazines N2(SiMe3)n(CO2SiMe3)4-n (n = 1,2) and recovered 70% of the titanium complex. The titanium complex would fixate nitrogen reductively using KC10H8 in 51% yield. Their method also determined reactivity of the Ti-N2-Ti complex with either tert-butyl isocyanate and (excess) phenylallene in 70% and 30% respectively. Both these pathways have not been investigated as in-depth as the carboxylation pathway and have no extraction from the metal complex reported.
8
Figure 5: reaction pathways of Kawaguchi and co-workers43
The methods by Chirik and co-workers and Kawaguchi and co-workers show an interesting approach towards the synthesis of substituted hydrazines. The recovery of the metal complex shows potency for catalytic employment of these pathways. Unfortunately, the yield for the fixation of nitrogen remains insufficient. These methods rely on either iodo- or chlorotrimethylsilane for the extraction of the product. Chlorotrimethylsilane is a fairly cheap chemical, whereas its iodo-counterpart is more costly, rendering the method by Kawaguchi more economical. Additionally, there is the tradeoff between regiospecificity and reaction time, where the regiospecific method can take up to 5 days. Overall, the method shows a reactive pathway where interesting and reactive hydrazine moieties are formed in decent yields. The main drawback for these methods is the nitrogen fixation, which proceeds in low yields and thus lowers the potential for turning these methods into catalytic cycles.
In 2019, Xi and co-workers reported the first rare-earth metal-promoted incorporation of N2 into organic compounds and provides a method for the synthesis on multiple hydrazine derivatives.9 Their discandium complex was able to methylate dinitrogen by adding potassium and methyl triflate several times to obtain the (N2Me2)2--bridged discandium complex (8) in 68% yield. Further reaction of this complex with a carbon-based electrophile produced the corresponding hydrazine derivative and regenerated the halogen-bridged discandium complex in 48-84% yield.
9
Figure 6: Hydrazine derivative synthesis by Xi and co-workers9
Figure 6 shows the versatility of this method. The different halogen-containing reactants lead to the regeneration of corresponding halogen-bridged discandium complex and have all been proven to yield the dinitrogen-discandium complex upon reacting with potassium and molecular nitrogen. This is a promising method due to its versatility, decent yield, and catalytic ability. This method does not only create a large variety of hydrazines, but also creates an azo compound upon reacting with iodine, which are commonly used as dyes. This methods shows a great first step towards a versatile catalytic cycle. It proceeds in decent to high yields and regenerates the scandium complex. Attempts to perform this methods in a one-pot fashion would bring this method closer to a catalytic cycle. Different triflates could also be explored to gain insight into generating different substituted hydrazines using this method.
10
Nitriles
Nitriles make interesting targets due to their wide array of follow-up reactions. They can easily be hydrolyzed to their corresponding acids and amides.44 They are susceptible to hydrogenation, carbocyanation, and are used as precursors for transition metal nitrile complexes.45–47 The carbon center of the nitrile is electrophilic, but also sufficiently acidic to form nitrile anions. Therefore nitriles can undergo both nucleophilic addition reactions and can alkylate a wide variety of electrophiles.48,49 Nitriles can therefore be of great value for application in synthetic chemistry.
Figure 7: Reactions using nitriles44–49
Cummins and co-workers formed a non-catalytic pathway towards a molybdenum cyanide complex in decent yields.50,51 The trisanilide molybdenum complex readily fixates nitrogen when exposed to a strong base (NaH). The resulting nitride complex would form the cyanide complex upon addition of MeOH2Cl and iPr3SiOTf, followed by deprotonation using LiHMDS and subsequentially adding SnCl2 and Me2NSiMe3. Release of HCN proved unsuccessful using a Lewis acid. However, it does demonstrate that reduction of molybdenum is possible through charge transfer from the ligand to metal which leads to imide deprotonation.27
De Vries and co-workers were the first to create an organic compound using the molybdenum nitride complex by Cummins.52 By reacting the complex with trifluoracetic anhydride, free amide CF3C(O)NH2 is formed in high yields. Unfortunately, this method is performed stoichiometrically and the complex is decomposed in the process.
In 2006, Cummins and co-workers managed to form the first synthetic cycle with this complex to form a variety of simple nitriles.53 The nitride complex first undergoes acylation with RC(O)Cl and SiMe3OTf to give the respective acylimide complex. After reduction using Mg/anthracene and Me3SiOTf to yield complex 10, the nitrile compound is released using a Lewis acid. The molybdenum complex is further reduced to regenerate nitride complex 9 and thus close the synthetic cycle. This cycle produces nitriles in up to 38% overall yield.
11
Figure 8: Nitrile and formamide synthesis using a molybdenum catalyst.50–53
The methods shown above demonstrate the excellent nitrogen fixating capabilities of molybdenum. Functionalization of the fixated nitrogen in a true catalytic fashion still remains difficult and proceeds in low yield with decomposition of the catalyst. To tackle this problem, Cummins and co-workers investigated the combined nitrogen fixation and functionalization capabilities of niobium and molybdenum complexes.54 Nitrogen could be fixated by the molybdenum complex, which in turn forms a heterodinuclear niobium/molybdenum N2 complex when exposed to complex 11. Upon reduction, the triple nitrogen bond was split, resulting in two metal complexes with a nitrogen atom on both metals. The resulting niobium complex is more reactive than the isoelectronic molybdenum complex.27 It can react with acyl chlorides to form free nitriles through oxo/nitride metathesis. The cycle is then closed using Tf2O and COCp2.55
12
Figure 9: Heterobimetallic nitrile synthesis by Cummins and co-workers.55
A variety of nitriles can be synthesized through this method in high yields. The method is pseudo-catalytic in regards to niobium, but is still stoichiometric for the molybdenum complex.
In 2016, Hou and co-workers managed to create a titanium species able to activate N2 by hydrogenolysis to form the multinuclear titanium imido complex 12.56 This complex is formed by adding N2 and H2 to Cp’Ti(CH2SiMe3)3 (Cp’=C5Me4SiMe3) at 180oC for 2 days (figure 10). Upon addition of an aromatic acid chloride, the corresponding aromatic nitrile is formed in a yield varying from 65-85%.13 Acid chlorides with electron withdrawing groups, as well as electron donating groups, proved to be suitable reagents. Remarkably, ammonia sensitive functional groups, such as chloromethyl and aldehyde groups, maintained their functionality in these conditions.
13 This method shows an interesting and efficient method for the production of a wide variety of nitriles. The complex is synthesized in an excellent yield and the nitriles are formed in decent yields. However, the necessity of trimethylsilyl methyllithium renders the method highly expensive. In addition, the use of a N2/H2 mixture still calls for the SMR as a necessity, which is the most energy-consuming step of the H-B process.
Over the past decade, Schneider and co-workers discovered and investigated the conversion of dinitrogen into nitriles using a rhenium(III) pincer complex [ReCl2(PNP)] (13, PNP=N(CH2CH2PtBu2)2).57– 60 The complex rapidly reduces to rhenium(V) nitride [Re(N)Cl(PNP)] with 1 equivalent of Na(Hg) in THF under N2 atmosphere with a 90% yield.59
Figure 11: Synthesis of the Rhenium complex by Schneider and co-workers59
Schneider’s first utilization of this complex yielded a synthetic cycle for the production of acetonitrile. By successive ethylation and deprotonation in benzene using EtOTf and KN(SiMe3)2, the 1-azavinylidene rhenium(III) complex [Re(N=CHCH3)Cl(PNP)] was formed in 81% yield. The addition of 2 equivalents of N-chlorosuccinimide resulted in the formation of acetonitrile in a >90% yield and gave the rhenium(I) [ReCl3(PNP)] complex, which could be reduced to 13 with 2 equivalents of Na(Hg) with a 70% yield.60
Further research showed different products could be obtained when employing various triflates. Their utilization of benzyl triflate (synthesized in situ from BnBr and AgOTf) yielded the corresponding benzonitrile with a decent yield of 57%.57
The most noteworthy achievement of Schneider and co-workers is their electrochemical reduction and photolytic splitting of dinitrogen (figure 12).58 Their rhenium catalyst was able to both be reduced and fixate nitrogen electrochemically in a yield of 69%. Bulk electrolysis of 14 at E= -1.65V resulted in full conversion and a 1H-NMR confirmed yield of 99%. After full conversion was observed, the electrolysis was continued at E= -1.85V to obtain 15 in 69% yield. Attempts were made using benzoic acid for the reduction as it is a byproduct of the nitrile synthesis, but better yields were obtained with 2,6-dichlorophenol (DCP), which has a conjugate base that is less prone to metal coordination.
In addition, this catalyst was able the split nitrogen under photolytic conditions (λ=390nm) with a yield of 95%. Excitation with a Xe lamp overcame the kinetically hindered N2-splitting by population of the N-N antibonding MOs. This photolysis reaction was able to proceed directly from the reaction mixture resulting from the electrolysis, although the yield was significantly lower (14% yield).
Upon addition of benzoyl chloride, three products were observed: benzamide (30%), and benzonitrile (64%) with equimolar benzoic acid. Respectively, benzonitrile and benzoic acid are products from the reaction of benzamide with benzoyl chloride, supporting that benzamide is the immediate product.58
14
Figure 12: Electrochemical reduction and photolytic splitting of dinitrogen by Schneider and co-workers57
Overall, this method is interesting due to the absence of any reactant in both the electrolytic nitrogen fixation and photolytic nitrogen splitting reactions, which makes this process incredibly economical. It generates both benzonitrile and benzamide, which are both useful targets in synthetic chemistry. The benzoyl chloride can be recycled by reacting the generated benzoic acid with benzotrichloride, reducing the total waste of this synthetic cycle.61 No purification is required after the photolysis step, but is a necessity after the electrolytic reduction and nitrogen fixation, impeding this chemical cycle from proceeding catalytically. Nevertheless, this method shows an economical and high-yielding approach for the synthesis of benzonitrile and could potentially be performed catalytically after optimization.
15
Silylamines
Silylated amines are interesting targets for a number of reasons. The N-Si bond is a useful linkage for masking primary and secondary amines. They function as a synthetic building block in various methods.62,63 They are used in the fabrication of silicon-nitride semiconductors in front-end electronic appliances, and are used to incorporate thermal resistance into ceramic materials as Si-N based polymers.64 Most importantly, silylated amines undergo facile hydrolysis to quantitatively yield ammonia.65 The main advantage of silylation over the H-B process is the absence of ammonia and dihydrogen, which can cause catalysts to suffer from poisoning.66
Silylation of dinitrogen has been studied intensively and several compounds have been found to transform N2 into silylamines. These compounds include transition metals as Ti, V, Cr, Mo, W, Fe, and Co.65 The most efficient catalysts found were Fe and Co catalysts, where the highest reported number of fixed nitrogen atoms is 320 per catalyst by Gagliardi and co-workers64 and 270 per transition metal by Masuda and co-workers15 The maximum yield found in these investigations is around a rather low amount of 50%, but are generally around 10-15%. This is most likely due to the formation of by-products such as disilanes formed via generation of silyl radicals.65
The method with the highest turnover number (TON) by Gagliardi and co-workers employed a dicobalt complex (figure 13) to transform N2 into trisilylamine with an initial TON of 195 [N(SiMe3)3]/[cat.]. Upon consecutive catalytic cycles, the TON could be increased to 320 [N(SiMe3)3]/[cat.]. However, a large excess of SiMe3Cl is necessary to achieve this TON with a N(SiMe3)3 yield of 25%.64 65% of the catalytic activity was retained in the second cycle (addition of 2000 eq. Me3SiCl and KC8). The dicobalt system gives a high TON at a low catalyst loading at 299 K. Employing KC8 as reducing agent accelerates the reaction 7-fold compared to K-metal, most likely because of the larger surfer area of the graphite. A consideration could be made between employing pure potassium instead of KC8 to slightly reduce the cost of reactants in trade for longer reaction times (95 h).
Figure 13: Catalytic cycle for N(SiMe3)3 using the dicobalt catalyst by Gagliardi and co-workers
The method by Masuda and co-workers works in a similar fashion: excess KC8 and SiMe3Cl converted N2 to N(SiMe3)3, but rather with a monocobalt catalyst. This catalyst has the best TON at low temperatures and long reaction times (233 K; 10 days). The total yield of this reaction is 27% N(SiMe3)3 and gives an excellent TON of 270, but is prohibitively slow to be useful in an industrial setting.
16 The methods by Masuda and co-workers and Gagliardi and co-workers both show excellent TONs, but low N(SiMe3)3 yields. Shifting focus towards efficiency of the reaction would ultimately lead to lower costs and less waste. The highest yield obtained so far by Murray and co-workers employed a Fe3Br3L catalyst (L = tris(β-diketiminate)cyclophane) to yield 50% N(SiMe3)3.67 Again, this method uses KC8 and SiMe3Cl as reactants under a nitrogen atmosphere in a low temperature (239 K). A total time of 96 h is needed to obtain this 50% yield, but a yield of 34% is already obtained after 24 h. This modest yield is mainly obtained due to the lower amount of SiMe3Cl and KC8 added.
The silylation methods shown above make use of KC8 as reducing agent. Currently, all reported methods make use of alkali metals as reducing agent (Li, Na, K, KC8), where formation of silyl radicals is unavoidable.65 Employing a more mild reducing agent could therefore aid in reducing the formation of disilanes and thus improve selectivity towards silylamines. The method previously mentioned for the formation of hydrazines by Chirik and co-workers (figure 4) shows liberation of N2-derived species can also occur electrophilic by silylation without the use of alkali metals.12,41
In 2016, Mézailles and co-workers performed a double hydrosilylation with 1,2-bis(dimethylsilyl)ethane and a tridentate phosphine molybdenum complex.68 The nitrogen fixation was performed reductively with stoichiometric amounts of sodium amalgam to obtain molybdenum complex 17. Upon heating this complex with an excess of 1,2-bis(dimethylsilyl)ethane, disilylamine 19 was formed by oxidative addition in 77% yield. When performed in a one-pot fashion, a yield of 57% was obtained, which is higher than the total yield of the two isolated steps (46%). This method shows a double silylation of dinitrogen that does not produce silyl radicals, preventing the formation of disilane. The group is currently investigating methods to transform complex 18 into 17 to create a chemical cycle and ideally create a catalytic system.
17
Heterocycles via N(SiMe
3)
3Mori and co-workers reported a one-pot method for the formation of heterocycles from dinitrogen via silylation, desilylation and Paal-Knorr type reactions.69 Their method used lithium as reducing agent on a fairly simple titanium complex (TiCl4 or Ti(OiPr4)) and SiMe3Cl to create N(SiMe3)3. The N(SiMe3)3 is then deprotected using CsF, which subsequentially reacts with (di)ketone moieties to form a variety of heterocycles and benzamide (figure 15). These heterocycle formations are most likely to proceed in a similar fashion as the Paal-Knorr synthesis.70
Figure 15: Synthesis of heterocycles by Mori and co-workers69
The variations in yield in the heterocycle synthesis shown above is mostly caused by steric hindrance. More bulky substrates result in lower yields. In the case of 20, better leaving groups such as ethyl carbonate and diethyl phosphonate resulted in higher yields. In this case, higher yields were also obtained using the TiOiPr4 complex instead of TiCl4. For the pathway for the synthesis of 21 using alkyne moieties, the yield is mostly determined by the inductive effects of the R-group. Electron withdrawing groups such as esters and nitriles result in the highest yields. All these pathways have also been proven to proceed using dry air instead of N2 gas, resulting in a slightly lower yield in most cases.71
Unfortunately, this method does not proceed catalytically due to the reactivity of fluoride with the SiMe3Cl. However, this entire method does not necessarily need to proceed catalytically. The catalytic segment of this method is mainly focused on the production of N(SiMe3)3, which is then applied in a stoichiometric fashion. Since the titanium complex is only used in the silylation of this method, the focus should shift to either performing the silylation step using SiMe3F or generating SiMe3Cl in the desilylation step. If obtainable, this method would then create two chemical cycles with catalytic SiMe3Cl and titanium catalysts.
18
Isocyanates
Isocyanates are organic compounds with the functional group R-N=C=O. Their electrophilic properties make them reactive toward a variety of nucleophiles. They can react with a urea group to obtain a biuret group, with amines to form urea groups, with H2O to form primary amines and with alcohols to form urethane groups. Isocyanates can undergo cyclization as dienophile in the Diels-Alder reaction and can form trimers when using aliphatic isocyanates.72
Figure 16: nucleophilic reactions of isocyanates
Isocyanates are currently formed by phosgenation of amines.73 They are used as insulation for construction, flexible foams used in furniture, weather resistant adhesives, and many other applications. Diisocyanates are key components in the synthesis of polyurethanes, which are used in a large variety of plastics.74 Currently, around 16 million tons of polyurethanes are produced worldwide annually.75
Figure 17: Isocyanate synthesis by phosgenation.
Chirik and co-workers managed to form an isocyanato bihafnocene μ-nitrido complex by exposing their earlier mentioned bihafnocene-N2 complex 4 in a frozen benzene solution to 1 eq. of CO, followed by warming to rt.19 The formed isocyanato bihafnocene complex (22) proved to be metastable and decomposed in a few hours. Complex 22 has shown to be very versatile and can undergo a variety of reactions (figure 18). Later research showed that terminal allenes and acetylenes would also add to the N=Hf bond of 22. Unfortunately, this method does not create a new C-N bond in this complex.76 All these methods show almost no release of product from the complex, with the exception of the methanediimine (23). In later attempts, Chirik and co-workers managed to generate N-ethyl urea with a yield of 79% by alkylation of complex 22 with EtOTf, followed by treatment with 4 equivalents of anhydrous hydrochloric acid.77 Unfortunately, they do not report any recovery of a (bi)hafnium complex.
19
Figure 18: Versatility of the isocyanato bihafnocene complex by Chirik and co-workers
In all these possible routes, only two products have been reported to be eliminated from the complex:
N-ethyl urea and the methanediimine. In all these reports, the complex reacts in a stoichiometric
fashion, rather than catalytic. The formation of the isocyanate ligand shows high speed, yield, and reactivity of the created complex, but without any catalytic potency, the applicability of this method is greatly reduced.
In 2017, Mazzanti and co-workers reported a diuranium-tripotassium complex with similar reactivity with CO as the method by Chirik and co-workers mentioned above.18 This method produces a terminal isocyanate at the uranium metals, but does not report any release of an isocyanate product. They do report the generation of highly toxic KCN as a byproduct, which makes this process more dangerous and less attractive.
Kawaguchi and co-workers managed to form isocyanatomethane as an intermediate in their preparation of a ureate niobium complex (figure 19).78 Their diniobium complex was able to fixate nitrogen without the use of any reducing agent. The resulting Nb-N2-Nb complex was efficiently methylated using MeI at 60 oC for 5 days. The complex was split into two identical imine niobium complexes using pyridine, enabling it to fixate CO2 to form the carbamate complex 25 by [2+2] cycloaddition. This complex immediately forms isocyanatomethane by extrusion, which then undergoes another [2+2] cycloaddition with the rest of the imine niobium complex 24 to form the N,N-dimethyl ureate niobium complex 26.
20
Figure 19: ureate niobium complex synthesis with isocyanatomethyl intermediate by Kawaguchi and co-workers78
The method discovered by Kawaguchi and co-workers contains no NMR spectroscopic evidence that the carbamate complex 25 and MeNCO are formed. This either implies that the formation of the ureate complex 26 is relatively fast compared to the consumption of CO2, or that the method proceeds through a different mechanism. Similar CO2 fixation reactions show that this methods is most likely to proceed through the mechanism mentioned above, meaning that the MeNCO does indeed react with
24 at a high pace.79–81
Following this research, Kawaguchi and co-workers created a synthetic cycle using vanadium complex
27 (figure 20).82 The complex is able to fixate nitrogen upon addition of 4 equivalents of KH to form a vanadium-nitrogen dimer, which can be oxidized by 1,4-benzoquinone to generate the vanadium monomer, cleaving the dinitrogen triple bond. The reaction of this complex with carbon monoxide led to the formation of a cyanate complex. Treatment of this complex with an alkyne produces an alkyne adduct and potassium cyanate. Solvation of the adduct in THF regenerates starting complex 27 and thus closes the synthetic cycle.
21
Figure 20: potassium isocyanate synthesis by Kawaguchi and co-workers82
The method above is a very promising method due to its high yields and excellent atom economy. The only formed byproduct is hydroquinone, which is easily reduced using sodium dichromate.83 Although this reduction of hydroquinone proceeds fast and with high yield, the necessary sodium dichromate is very toxic and is produced at high temperatures (1000 oC).84 Therefore, sodium dichromate loses appeal for its application in catalytic systems.
This method makes uses of mostly volatile reagents which, in combination with high yields, facilitates easy purification. The reduction step with KH, followed by oxidation with 1,4-benzoquinone and the production of the alkyne adduct are the only steps that require purification. Finetuning of this method and its purification could lead to a catalytic cycle for efficient synthesis of KNCO.
In 2015, Sita and co-workers created a versatile chemical cycle wherein a molybdenum complex generates Me3SiNCO in four steps.85 In the first step, the complex fixates dinitrogen reductively upon addition of Na0 and N2. In the second step, a terminal imido moiety is formed by photolytic N≡N bond cleavage and N-atom functionalization with Me3SiCl. This step produces the starting complex 28 as a byproduct. The formed terminal imido complex forms Me3SiNCO through simultaneous oxygen-atom transfer and nitrene group transfer upon addition of CO2. The molybdenum complex 28 can be regenerated through oxygen-atom transfer with Me3SiCl and forms disiloxane as a byproduct. Complex
22
Figure 21: Chemical cycle for the synthesis of isocyanates by Sita el al.85
The method shown above shows an efficient chemical cycle for the production of isocyanates from dinitrogen under somewhat mild conditions. The cycle has a great atom-economy due to the only byproducts being NaCl and disiloxane, which is a commonly used sealant for paints, inks, and mainly cosmetics.86 The method has a negative greenhouse gas emission due to the usage of CO2. The main drawbacks of this method are the reaction time, which in total takes up more than 5 days, and the necessity of 2 equivalents of sodium per complex 28. Nevertheless, this cycle produces a product that could be very interesting in organic synthesis due to its electrophilic character and ability to hydrolyze to ammonia. Refinements of this cycle should be investigated to let the cycle proceed in a true catalytic fashion.
23
Oxamide
The method by Chirik and co-workers for the creation of an isocyanate ligand (figure 18) has also been shown to create substituted oxamide moieties when higher CO pressure is applied.87 The complex then easily adds alkyl halides which, after protonolysis, result in free N,N’-disubstituted oxamides. These products are useful precursors for the preparation of various heterocycles, N-heterocyclic carbenes, and N,N’-diamines.87 Addition with primary and secondary silanes and terminal alkynes was also accomplished.
Figure 22: Synthesis of oxamide structures by Chirik and co-workers87
This method shows an efficient and versatile method for the production of oxamide moieties with yields varying from 80-94% and reaction times varying from 1-16 hours. The reaction proceeds in a stoichiometric fashion. This pathway results in the dichloro hafnium complex 30, which can transformed in the precursor for the bihafnium complex shown in Chirik’s hydrazine synthesis (figure 4). This iodination used boron triiodide in excess, which leaves this route uninteresting for catalysis due to the very high reagent costs. Efficient nitrogen fixation from 30 to create the hafnium complex
4 would therefore greatly increase the possibility of these methods to be performed catalytically.
24
Diazonium salts
Another interesting and reactive target is the diazonium salt. Diazonium salts are strong electron-withdrawing R-N2+X- groups that can be used as intermediates for a large variety of compounds, the most interesting being azo compounds, aryl halides and benzonitrile.61,88 They can also react with phosphines as nitrogen-based Lewis acids for tunable synthesis of azophosphonium salts.89–91 They are traditionally prepared from aniline, NaNO2 and HCl and are rarely isolated due to their lack of stability at room temperature, which is why they are often prepared in situ or isolated as the more stable BF4 -or SO42- salts.92
In 2006, Winkler and co-workers showed a synthesis of benzenediazonium ions through cryogenic nitrogen fixation.93 By slow sublimation of an aryl halogen with a large excess of argon at 8 K combined with continuous argon excitation by microwave, the phenyl cation was formed in low yields.94 These cations form benzenediazonium ions when exposed to nitrogen (0.5 – 1.0 % in argon). This method was monitored by IR-spectroscopy and does not report any isolated product. Application of this method for synthetic purposes is still in its infancy, but it does lay ground for a whole new method of nitrogen fixation without the use of any reducing agents.
25
C-H activation
Recently, Holland and co-workers have demonstrated the cross-coupling opportunities of C-H bonds with dinitrogen to form silylated aniline.95 Their iron complex coordinates with benzene when reduced with sodium and 15-crown-5 (15c5).96 Upon continuous exposure to an excess of sodium, N2, and SiMe3Br, silylated aniline is formed with a yield of 92% and N(SiMe3)3 is formed with 135% yield in regards to Fe. Multiple additions of Na and SiMe3Br suggests this method proceeds in a catalytic fashion, but suffers a substantial loss of active species throughout the cycle. This is the first reported cross-coupling of C-H bonds with dinitrogen.
Figure 25: Cross-coupling of benzene with dinitrogen by Holland and co-workers95
This methods uses 6 equivalents of sodium and 15c5 for both reduction of the catalyst and activation of the silyl group, rendering Na(15c5)H as a byproduct. Both products undergo facile hydrolysis towards ammonia and aniline. This method provides an interesting pathway that maps a route towards future catalytic systems.
26
Nitrogen fixation at boron
Nitrogen fixation and functionalization tends to focus mostly on transition metal catalysts. These metals bind dinitrogen using both empty and filled d orbitals by σ-donation and π-backdonation. Borylene can perform similar σ-donation and π-backdonation using its empty sp2 and filled p orbital.97 Several studies have investigated the reactivity of borylene with nitrogen, and a few nitrogen containing products have been found.98–101
Figure 26: Orbital interaction for N2 activation using transition metals (left) and borylenes (right)97
In 2018, Braunschweig and co-workers lay ground for dinitrogen activation using (CAAC)BBr2Dur.99 The boron atom is reduced similarly to transition metals using KC8. The reduced complex intermediate activates dinitrogen to yield complex 31 in 64% yield. This complex can further react to form the dinitrogen complex 32 under ambient air or the diradical hydrazino complex 33 in purified water. There are many [(CAAC)BR2]• radicals known to be stable, which have a similar stability as complex 33.102
27 A year later, Braunschweig and co-workers discovered that employing the bulkier ligand 1,3,5-triisopropylbenzene (Tip) on the boron group results in the formation of a tetrazene moiety, rather than the hydrazine moiety (figure 28).98 Tetrazene compounds have been of considerable interest due to their high energetic performance and challenging synthesis.103
Figure 28: Diboratetrazene synthesis by Braunschweig and co-workers.98
The methods shown above do not show any release of the hydrazine or tetrazene moieties. Generally, boron-nitrogen bonds can be cleaved using a NaOH/H2O2 mixture at low temperatures.104 However, the resulting boron compound requires halogenation to regenerate the starting compound. These pathways show fixation and functionalization of nitrogen is feasible using borylene moieties as catalyst. Nitrogen fixation at boron will therefore remain an interesting topic with a surplus of potential pathways yet to be discovered.
Lithium
Previously, Mori and co-workers have shown lithium can be applied as a reducing agent for their TiCl4 catalyst for the synthesis of heterocycles. Among the main group elements, lithium is exceptional because it slowly reacts with N2 at room temperature, ultimately leading to lithium nitride (NLi3)n, which can be used as a hydrogen storage medium. This exceptional facile dinitrogen activation, compared to other alkali metals, is due to the fact that lithium nitrides have the highest atomization energy, the shortest M-N bond distance, and the largest M-N charge separation as well as interaction energy.105 A minimum of eight lithium atoms are necessary for cleaving the triple bond of dinitrogen in a highly exothermic fashion. The resulting lithium nitride clusters are readily hydrolyzed to form NH3 and LiOH. While homoatomic lithium clusters are unable to give N≡N bond separation catalytically, incorporating other elements in the clusters could provide promising strategies for nitrogen fixation catalyst design.106
Some progress has been made into creating catalysts from superalkali lithium species. Superalkalis are clusters of atoms that exhibit the properties of a single alkali metal. They are a subclass of superatoms.107
In 2017, Meloni and co-workers investigated the catalytic properties of the superalkali species Li3F2. Increasing the Li3F2 units up to 6 to give (Li3F2)nN2, where n=1-6, gave complete dissociation of the N≡N bond. This was confirmed through visualized MOs, bond length and bond order computations.107
28
Figure 29: Geometries of (Li3F2)nN2 clusters, calculated by Meloni and co-workers107
The work by Meloni and co-workers does not apply their clusters for functionalization of nitrogen, but it does show activation of dinitrogen can be achieved using superalkali species.
In 2018, Goel and co-workers investigated the nitrogen fixating capabilities of the superalkali species BLin (n=5-7). Their calculations suggested that BLi6 and BLi7 bind N2 in a side-on fashion with a binding energy of -18.17 kcal/mol (BLi6) and -18.90 kcal/mol (BLi7). Of these two species, BLi6 has the lowest ionization energy (3.84 eV), making it lower than the most electropositive element Cs (3.90 eV). BLi6 can therefore be considered a superalkali with a high charge transfer capacity.106 By DFT calculations, they found evidence that the BLi6 cluster is an efficient and active catalyst for N2 fixation towards NH3 without destroying its structural integrity. However, this method requires the cluster to be deposited on a substrate for experimental application. Both graphene and BN-sheet have been explored, in which BN-sheet stabilizes the BLi6 cluster without altering its activity towards N2.
These calculations show that superalkali lithium species have great potential in nitrogen fixation and could eventually lead to efficient catalysts. Although the nitrogen fixating capabilities of lithium has been known for over half a century, research towards its catalytic application is still in its infancy.108
29
Discussion
Nitrogen fixation for the synthesis of Nr beyond ammonia has made decent progress throughout the year, the most notable compounds being hydrazines, nitriles, isocyanates, and silylamines. Out of these processes, the synthesis of silylamines has been found to be the pathway with the most catalytic properties so far.
The synthesis of silylamines knows a variety of catalytic pathways, with TONs up to 320 moles per catalyst and total yields up to 50% N(SiMe3)3 while employing cheap catalysts consisting of 3d metals. The by-products in these pathways are mostly chlorinated reducing agents and disilane. However, the main application of trisilylamine would be to hydrolyze it to ammonia, meaning the formation of 3 equivalents of trimethylsilanol as by-product. So, even though there are different fields that can still be explored for this pathway (mild reducing agents, electrophilic SiMe3Cl attack), the method will only have industrial applicability when it at least performs at the same level as the H-B process.
Hydrazines have been reported to be synthesized with scandium, titanium and hafnium complexes. All of these methods run in full chemical cycles, but so far no catalytic system has been reported. These methods show great versatility for the synthesis of different hydrazines. Xi and co-workers’ discandium complex can form a variety of substituted hydrazines depending on the halogen-containing reactant and Kawaguchi and co-workers’ titanium complex has shown to react with tert-butyl isocyanate and phenylallene. Currently, hydrazine is formed in an eco-friendly manner from ammonia. The main asset of these methods over the current hydrazine synthesis is the ability to form a large variety of substituted hydrazines with controlled regioselectivity and will be very interesting if performed catalytically.
Nitriles have been shown to be formed with their corresponding acid chloride and either a titanium, molybdenum/niobium or rhenium complex. Both the method by Cummins (Mo/Nb) and Hou (Ti) require three electrons per N-atom for regeneration of the metal. This has been achieve by the use of external reducing agents, H2 for Hou and alkali metals for Cummins. Hou’s titanium complex utilizes a N2/H2 mixture for the nitrogen fixation step. This mixture is formed through the SMR reaction and would therefore be of low value as an alternative for the H-B process. The method by Schneider and co-workers shows an interesting new approach by electrochemical nitrogen fixation and photolytic splitting, diminishing the need for a reducing agent. In this method, nitriles are formed through their amide intermediate in high yields. The regeneration of the metal requires only two electrons for each step. However, no catalytic cycle has been produced yet. Nitriles can be of great value in organic synthesis and having an economical catalytic method to form these structures is of great interest. Isocyanates have been shown to be useful ligands in the formation of various nitrogen-containing complexes. Their electrophilic nature renders elimination from the complex without decomposition difficult. So far, only two isocyanate products have been formed: potassium isocyanate and (trimethylsilyl)isocyanate. Currently, the leading appliance for isocyanates is as building block for polyurethane synthesis wherein these products have little applicability. However, the (trimethylsilyl)isocyanate can be hydrolyzed to ammonia and be used in the formation of urea, both main components in fertilizer, and the potassium isocyanate is largely used in organic synthesis of, for example, herbicides. The worldwide production of the salt was 20,000 tons in 2006.109 Sita and co-worker showed elimination is possible upon addition of CO2 to a monosilylated nitrogen-molybdenum complex to create (trimethylsilyl)isocyanate. This route shows a similar intermediate as the silylation method by Mézailles, where a molybdenum complex creates a cyclic disilylamine. Possibly, new pathways to isocyanates can be discovered by combining silylation methods with Sita’s isocyanate
30 synthesis, performing the silylation method in a N2/CO2 mixture. This would lead to less trimethylsilanol as by-product in regards to N(SiMe3)3 production.
The bihafnium complex developed by Chirik and co-workers has shown to be capable of generating nitriles, ureas, diimides and oxamides. Additionally, the complex shows high versatility when employing an isocyanato ligand. Generally, their methods result in decent to high yields starting from the bihafnium dinitrogen complex. The main challenge in their endeavors is the fixation of nitrogen step, which has a highest reported yield of 27%. The nitrogen fixation step starting with the diiodo hafnium complex would make most chemical cycles dependent on SiMe3I for the elimination step, rather than the less costly SiMe3Cl. If formed, the dichlorohafnium complex can be iodinated with triiodoborane (figure 23), but results in an even more expensive pathway with an extra reaction step. The largest area of interest for this method would be to perform the nitrogen fixation starting from the dichlorohafnium complex, rather than its iodo counterpart.
Conclusions and outlook
Nitrogen fixation and functionalization with transition metals is a difficult and important challenge in organometallic chemistry. Over the past decade, reasonable progress has been made towards generating chemical cycles, most notably nitriles, hydrazines, isocyanates and silylamines. Unfortunately, the amount of catalytic methods for carbon-nitrogen bond formation is low. The aforementioned pathways show nitrogen fixation is not always the challenge. Breaking the nitrogen-metal bonds without decomposition proves to be an equally difficult assignment and is usually accomplished with a halogen-containing trimethylsilane.
The most successful area is the catalytic silylation of dinitrogen. A large variety of metals have proven to be suitable catalysts for the formation of N(SiMe3)3 from SiMe3Cl and a reducing agent. The challenge remains to perform these methods without the use of strong reducing agents. Currently, silylamines are commonly hydrolyzed to obtain ammonia, which is nowhere near sensible as an alternative for the H-B process.
Both electro- and photochemical methods lack research for nitrogen fixation. A few highly efficient methods have been investigated and show promise for future methods. Additionally, bimetallic catalyst have been underdeveloped as well. Not only do bimetallic catalysts combine properties related to the two individual metals, they also generate new and distinct properties due to synergistic effects between the two metals present.110
Transition metal chemistry has been the main focus for nitrogen fixation and functionalization. While this area is large, interesting, and full of potential, other types of atoms have also shown its ability to fixate and functionalize nitrogen. Both boron complexes and superalkali lithium clusters have proven their reactivity with diatomic nitrogen. These methods are largely underdeveloped and could potentially aid in understanding nitrogen fixation and contribute in its catalyst design.
31
References
1. Denk TRA, Mohn J, Decock C, et al. The nitrogen cycle: A review of isotope effects and isotope modeling approaches. Soil Biol Biochem. 2017;105:121-137. doi:10.1016/j.soilbio.2016.11.015 2. Fowler D, Coyle M, Skiba U, et al. The global nitrogen cycle in the twenty- first century. R Soc.
2013;368. doi:10.1098/rstb.2013.0164
3. Smil V. Detonator of the population explosion. Nature. 1999;400:415. doi:10.1038/22672 4. Kyriakou V, Vasileiou E, Stoukides M, et al. An Electrochemical Haber-Bosch Process An
Electrochemical Haber-Bosch Process. Joule. 2019:1-17. doi:10.1016/j.joule.2019.10.006 5. Dong Y, Xu L, Yang Z, Zheng H, Chen L. Aggravation of reactive nitrogen flow driven by human
production and consumption in Guangzhou City China. Nat Commun. 2020;11(1209):1-10. doi:10.1038/s41467-020-14699-x
6. Stevens C. Nitrogen in the environment. Science (80- ). 2019;363(6427):578-580. doi:10.1126/science.aav8215
7. Chen JG, Crooks RM, Seefeldt LC, et al. Beyond fossil fuel–driven nitrogen transformations.
Science (80- ). 2018;360(6391). doi:10.1126/science.aar6611
8. Bezdek MJ, Chirik PJ. Expanding Boundaries: N2 Cleavage and Functionalization beyond Early Transition Metals. Angew Chemie - Int Ed. 2016;55(28):7892-7896.
doi:10.1002/anie.201603142
9. Lv ZJ, Huang Z, Zhang WX, Xi Z. Scandium-Promoted Direct Conversion of Dinitrogen into Hydrazine Derivatives via N-C Bond Formation. J Am Chem Soc. 2019;141(22):8773-8777. doi:10.1021/jacs.9b04293
10. Xue X, Chen R, Yan C, et al. Review on photocatalytic and electrocatalytic artificial nitrogen fixation for ammonia synthesis at mild conditions : Advances , challenges and perspectives.
Nano Res. 2019;12(1):1-21. doi:10.1007/s12274-018-2268-5
11. Folkesson B, Larsson R. Nitrogen fixation. I. Synthesis of Hydrazine from molecular nitrogen by transition metal complexes. Acta Chem Scand. 1979;(33):347-357.
12. Knobloch DJ, Toomey HE, Chirik PJ. Carboxylation of an ansa-Zirconocene Dinitrogen Complex : Regiospecific Hydrazine Synthesis from N 2 and CO 2. J Am Chem Soc. 2008;(130):4248-4249. doi:10.1021/ja8008506
13. Guru MM, Shima T, Hou Z. Conversion of Dinitrogen to Nitriles at a Multinuclear Titanium Framework. Angew Chemie - Int Ed. 2016;55(40):12316-12320. doi:10.1002/anie.201607426 14. Klopsch I, Kinauer M, Finger M, Würtele C, Schneider S. Conversion of Dinitrogen into
Acetonitrile under Ambient Conditions Angewandte. Angew Chemie - Int Ed. 2016;55:4786-4789. doi:10.1002/anie.201600790
15. Suzuki T, Fujimoto K, Takemoto Y, Wasada-tsutsui Y, Ozawa T. Efficient Catalytic Conversion of Dinitrogen to N(SiMe3)3 Using a Homogeneous Mononuclear Cobalt Complex. ACS Catal. 2018;8(4):3011-3015. doi:10.1021/acscatal.7b04351
16. Ghana P, Krüchten F, Spaniol T, Leusen J, Kögerler P, Okuda J. Conversion of dinitrogen to tris(trimethylsilyl)amine catalyzed by titanium triamido-amine complexes †. Chem Commun
(Camb). 2019;(55):3231-3234. doi:10.1039/c8cc09742a
17. Piascik AD, Li R, Wilkinson HJ, Green JC, Ashley AE. Fe-Catalyzed Conversion of N 2 to N(SiMe 3 ) 3 via an Fe-Hydrazido Resting State. J Am Chem Soc. 2018;(140):10691-10694.
32 doi:10.1021/jacs.8b06999
18. Falcone M, Chatelain L, Scopelliti R, Živković I, Mazzanti M. Nitrogen reduction and
functionalization by a multimetallic uranium nitride complex. Nat Publ Gr. 2017;547:332-335. doi:10.1038/nature23279
19. Semproni SP, Chirik PJ. Synthesis of a Base-Free Hafnium Nitride from N2 Cleavage: A
Versatile Platform for Dinitrogen Functionalization. J Am Chem Soc. 2013;(135):11373-11383. doi:10.1021/ja405477m
20. Lv Z, Wei J, Zhang W, et al. Direct transformation of dinitrogen: synthesis of N-containing organic compounds via N−C bond formation. Natl Sci Rev. 2020:1-59.
doi:10.1093/nsr/nwaa142
21. Bozso F, Ertl G, Grunze M, Weiss M. Interaction of Nitrogen with Iron Surfaces. J Catal. 1977;41:18-41. doi:10.1016/0021-9517(77)90237-8
22. Li C, Wang T, Gong J. Alternative Strategies Toward Sustainable Ammonia Synthesis. Trans
Tianjin Univ. 2020;26(2):67-91. doi:10.1007/s12209-020-00243-x
23. Appl M. Ammonia, Ullmann’s Encyclopedia of Industrial Chemistry.; 2006.
24. van der Ham CJM, Koper MTM, Hetterscheid DGH. Challenges in reduction of dinitrogen by proton and electron transfer. chem soc rev. 2014;43(15):5183-5191. doi:10.1039/c4cs00085d 25. Kim H, Doan VD, Cho WJ, et al. Intriguing Electrostatic Potential of CO : Negative Bond-ends
and Positive Bond-cylindrical-surface. Sci Rep. 2015;5(1):1-9. doi:10.1038/srep16307
26. Kim S, Loose F, Chirik PJ. Beyond Ammonia: Nitrogen−Element Bond Forming Reactions with Coordinated Dinitrogen. Chem Rev. 2019;120(12):5637-5681.
doi:10.1021/acs.chemrev.9b00705 27. Nishibayashi Y. Nitrogen Fixation.; 2017.
28. Fomitchev D V., Bagley KA, Coppens P. The First Crystallographic Evidence for Side-On Coordination of N2 to a Single Metal Center in a Photoinduced Metastable State. J Am Chem
Soc. 2000;122(3):532-533. doi:10.1021/ja993623p
29. Walter MD. Recent Advances in Transition Metal-Catalyzed Dinitrogen Activation. In:
Advances in Organometallic Chemistry. Vol 65. 1st ed. Elsevier Inc.; 2016:261-377.
doi:10.1016/bs.adomc.2016.03.001
30. Pierpont AW, Cundari TR. Dinitrogen activation by low-coordinate transition metal complexes.
J Coord Chem. 2011;64(18):3123-3135. doi:10.1080/00958972.2011.616586
31. Chatt J, Hussain W, Leigh GJ, Ali HM, Pickett CJ, Rankin DA. The Preparation and Properties of Some Diphosphines R2PCH2CH2PR2 (R = Alkyl or Aryl) and of their Rhenium(I) Dinitrogen Derivatives. Dalt Trans. 1985;(6):1131-1136. doi:10.1039/DT9850001131
32. Hussain W, Leigh GJ, Ali HM, Pickett CJ, Rankin DA. Dinitrogen Binding and Electrochemistry in Complexes of Molybdenum and Tungsten. Dalt Trans. 1984;(8):1703-1708.
doi:10.1039/DT9840001703
33. Mackay BA, Fryzuk MD. Dinitrogen Coordination Chemistry: On the Biomimetic Borderlands.
Chem Rev. 2004;104(2):385-402. doi:10.1021/cr020610c
34. Ho BM, Lukoyanov D, Yang Z, Dean DR, Seefeldt LC. Mechanism of Nitrogen Fixation by Nitrogenase: The Next Stage. Chem Rev. 2014;114(8):4041-4062. doi:10.1021/cr400641x
33 35. Laplaza CE, Cummins CC. Dinitrogen Cleavage by a Three-Coordinate Molybdenum ( III )
Complex. Science (80- ). 1995;268(5212):861-863. doi:10.1126/science.268.5212.861 36. Chu W, Wu C, Hsu H, Cheng N, Uni K. Catalytic Reduction of Hydrazine to Ammonia by a
Vanadium Thiolate Complex. Inorg Chem. 2006;45(8):3164-3166. doi:10.1021/ic960098b 37. Schrock RR, Glassman TE, Vale MG. Cleavage of the N-N Bond in a High-Oxidation-State
Tungsten or Molybdenum Hydrazine Complex and the Catalytic Reduction of Hydrazine. J Am
Chem Soc. 1991;113(2):725-726. doi:10.1021/ja00002a079
38. van Tamelen EE, Fechter RB, Schneller SW. Conversion of molecular nitrogen to hydrazine. J
Am Chem Soc. 1969;91(25):7196-7196. doi:10.1021/ja01053a054
39. Sridhar S, Srinivasan T, Virendra U, Khan AA. Pervaporation of ketazine aqueous layer in production of hydrazine hydrate by peroxide process. Chem Eng J. 2003;94:51-56. doi:10.1016/S1385-8947(03)00045-7
40. Shilov AE. Fixation of Nitrogen in Solution in the Presence of Transition Metal Complexes. Russ
Chem Rev. 1974;43(5):378-398. doi:10.1070/RC1974v043n05ABEH001811
41. Bernskoetter WH, Lobkovsky E, Chirik PJ. Nitrogen – Carbon Bond Formation from N2 and CO2 Promoted by a Hafnocene Dinitrogen Complex Yields a Substituted Hydrazine. Angew Chemie
- Int Ed. 2007;46:2858-2861. doi:10.1002/anie.200604099
42. Bernskoetter WH, Olmos A V, Lobkovsky E, Chirik PJ. N2 Hydrogenation Promoted by a Side-On Bound Hafnocene Dinitrogen Complex. Organometallics. 2006;25(13):1021-1027. doi:10.1021/om0509037
43. Nakanishi Y, Ishida Y, Kawaguchi H. Nitrogen–Carbon Bond Formation by Reactions of a Titanium– Potassium Dinitrogen Complex with Carbon Dioxide, tert-Butyl Isocyanate, and Phenylallene. Angew Chemie. 2017;129(31):9321-9325. doi:10.1002/anie.201704286
44. Kukushkin VY, Pombeiro AJL. Metal-mediated and metal-catalyzed hydrolysis of nitriles. Chim
Acta. 2005;358(1):1-21. doi:10.1016/j.ica.2004.04.029
45. Rach SF, Kühn FE. Nitrile Ligated Transition Metal Complexes with Weakly Coordinating Counteranions and Their Catalytic Applications. Chem Rev. 2009;109(5):2061-2080. doi:10.1021/cr800270h
46. Nakao Y, Yada A, Ebata S, Hiyama T. A Dramatic Effect of Lewis-Acid Catalysts on Nickel-Catalyzed Carbocyanation of Alkynes. J Am Chem Soc. 2007;129(9):2428-2429.
doi:10.1021/ja067364x
47. Barrault J, Pouilloux Y. Synthesis of fatty amines . Selectivity control in presence of multifunctional catalysts. Catal Today. 1997;37(2):137-153. doi:10.1016/S0920-5861(97)00006-0
48. Smith AL, Tan P. Creatine Synthesis : An Undergraduate Organic Chemistry Laboratory Experiment. J Chem Educ. 2006;83(11):1654-1657. doi:10.1021/ed083p1654
49. Denmark SF. Organic Reactions, Volume 100.; 2019.
50. Curley JJ, Cozzolino AF, Cummins CC. Nitrogen fixation to cyanide at a molybdenum center.
Dalt Trans. 2011;40:2429-2432. doi:10.1039/c0dt01326a
51. Laplaza CE, Johnson MJA, Peters JC, et al. Dinitrogen Cleavage by Three-Coordinate Molybdenum ( III ) Complexes : Mechanistic and Structural Data. J Am Chem Soc. 1996;118(36):8623-8638. doi:10.1021/ja960574x
34 52. Henderickx H, Kwakkenbos G, Peters A, Spoel J Van Der, Vries K De. Direct formation of an
organonitrogen compound from a molybdenum nitrido species. chem commun. 2003;(16):2050-2051. doi:10.1039/B305774G
53. Curley JJ, Sceats EL, Cummins CC. A Cycle for Organic Nitrile Synthesis via Dinitrogen Cleavage.
J Am Chem Soc. 2006;128(43):14036-14037. doi:10.1021/ja066090a
54. Mindiola DJ, Meyer K, Cherry JF, Baker TA, Cummins CC. Dinitrogen Cleavage Stemming from a Heterodinuclear Niobium / Molybdenum N 2 Complex : New Nitridoniobium Systems Including a Niobazene Cyclic Trimer. Organometallics. 2000;19:1622-1624.
doi:10.1021/om000159k
55. Figueroa JS, Piro NA, Clough CR, Cummins CC. A Nitridoniobium ( V ) Reagent That Effects Acid Chloride to Organic Nitrile Conversion : Synthesis via Heterodinuclear ( Nb / Mo ) Dinitrogen Cleavage , Mechanistic Insights , and Recycling. J Am Chem Soc. 2006;128(3):940-950. doi:10.1021/ja056408j
56. Shima T, Hu S, Luo G, Kang X, Luo Y, Hou Z. Dinitrogen Cleavage and Hydrogenation by a Trinuclear Titanium Polyhydride Complex. Science (80- ). 2013;340:1549-1553.
doi:10.1126/science.1238663
57. Klopsch I, Schendzielorz F, Volkmann C, Würtele C, Schneider S. Synthesis of Benzonitrile from Dinitrogen. Z Anorg Allg Chem. 2018;644:916-919. doi:10.1002/zaac.201800181
58. Schendzielorz F, Finger M, Abbenseth J, Christian W, Krewald V, Schneider S. Metal-Ligand Cooperative Synthesis of Benzonitrile by Electrochemical Reduction and Photolytic Splitting of Dinitrogen. Angew Chemie - Int Ed. 2019;58(2):830-834. doi:10.1002/anie.201812125
59. Klopsch I, Finger M, Wu C, Milde B, Werz DB, Schneider S. Dinitrogen Splitting and
Functionalization in the Coordination Sphere of Rhenium. J Am Chem Soc. 2014;136(19):6881-6883. doi:10.1021/ja502759d
60. Klopsch I, Kinauer M, Finger M, Würtele C, Schneider S. Conversion of Dinitrogen into Acetonitrile under Ambient Conditions. Angew Chemie - Int Ed. 2016;55:4786-4789. doi:10.1002/anie.201600790
61. Maki T. Benzoic Acid and Derivatives. In: Ullmann’s Encyclopedia of Industrial Chemistry. ; 2012:329-341. doi:10.1002/14356007.a03
62. Rhiel M, Weller F, Pebler J. [SbN(SbCl)3(NSbCl2)(NSiMe3)2·SbCl3], an Unusual
Nitridoantimony Complex with a Heterocubane Structure. Angew Chemie. 1994;106(5):599-600. doi:10.1002/anie.199405691
63. Okusu S, Hirano K, Tokunaga E, Shibata N. Organocatalyzed Trifluoromethylation of Ketones and Sulfonyl Fluorides by Fluoroform under a Superbase System. ChemistryOpen. 2015;4:581-585. doi:10.1002/open.201500160
64. Siedschlag RB, Bernales V, Vogiatzis KD, et al. Catalytic Silylation of Dinitrogen with a Dicobalt Complex. J Am Chem Soc. 2015;137(14):4638-4641. doi:10.1021/jacs.5b01445
65. Tanabe Y, Nishibayashi Y. Recent advances in catalytic silylation of dinitrogen using transition metal complexes. Coord Chem Rev. 2019;389:73-93. doi:10.1016/j.ccr.2019.03.004
66. Dzik WI. Silylation of of Dinitrogen Dinitrogen Catalyzed Catalyzed by by Silylation. Inorganics. 2016;4(3):21-29. doi:10.3390/inorganics4030021
67. Ferreira RB, Cook BJ, Knight BJ, Catalano VJ, Garc R, Murray LJ. Catalytic Silylation of Dinitrogen by a Family of Triiron Complexes. Am Chem Soc. 2018;8:7208-7212.
