Synthesis and Characterization of Homogeneous and Surface Attached Catalysts for Small Molecule Activation

Synthesis and Characterization of Homogeneous and Surface Attached Catalysts for Small Molecule Activation Dissertation submitted to the Faculty of Mathematics and Natural Sciences Institute of Inorganic Chemistry Christian-Albrechts-University of Kiel for the degree of Dr. rer. nat. Submitted by Sven Froitzheim Kiel, 2024 1. Gutachter: Prof. Dr. F. Tuczek 2. Gutachter: Prof. Dr. C. Näther Tag der mündlichen Prüfung: 08.11.2024 Zum Druck genehmigt: 08.11.2024 Der Dekan; Prof. Dr. Frank Kempken I This Thesis was written at the Institute of Inorganic Chemistry of the Christian-Albrechts-University of Kiel under the supervision of Prof. Dr. Felix Tuczek between March 2018 and October 2023. III Acknowledgement Als erstes möchte ich mich bei meinem Doktorvater Prof. Dr. Felix Tuczek für die Möglichkeit meine Doktorarbeit in seinem Arbeitskreis anzufertigen bedanken. Neben den spannenden und vielseitigen Themen in der Forschung, gelang es uns ein paar sehr hochwertige und lehrreiche Vorlesungen in den drei Coronasemestern zu gestalten. Herrn Prof. Dr. Christian Näther danke ich für die Übernahme des Zweitgutachtens, dem Lösen der Einkristallstrukturen und die Mitarbeit an den Publikationen. Ein besonderer Dank geht auch an Dr. Tobias Engesser. Danke für die vielen Stunden der Diskussion über Spektren, Rechnungen und unverstandene Ergebnisse, die Unterstützung bei der zweiten Publikation, das Korrekturlesen der Arbeit und wenn es nötig war - das Stubsen in die richtige Richtung. Dr. Jan Krahmer danke ich für die vielen und teilweise sehr langen NMR-Messungen. Auch danke ich dir für die Starthilfe bei der Elektrochemie. Das Entenzeichnen hat Spaß gemacht. Nicolas Le Poul danke ich für die tolle Kooperation, den Wissensaustausch und die intensive Betreuung während unseres Aufenthalts in Brest. Danke auch für die Aufnahme jenen Fotos, welches die Basis für das Frontcover unserer Publikation bildet. Mein Dank geht auch an Stephanie Pehlke und Jaqueline Pick für die Aufnahme der Elementarund Halogenidanalysen. Dem Arbeitskreis Tuczek danke ich für die tolle Zusammenarbeit und Stimmung während meiner Zeit. Die Aufenthalte in Sehlendorf gehören dabei definitiv zu meinen Highlights. Ein besonderer Dank geht auch an die Stickstofffixierung und vor allem an Jannik Junge. Danke für 5 Jahre mit viel Spaß, wertvollem wissenschaftlichen Austausch und einer tollen Zusammenarbeit - ohne dich hätte es nur halb so viel Spaß gemacht. Ein großer Dank geht auch an meine Kommolitonen und Freunde, die mich über die vielen Jahre begleitet haben. Christoph Bohl, Daniel Hugenbusch, Marc Lehr, Jonas van Dinter, Jannik Junge, Jannik Benecke, Tobias Haase, Felix Hartmann, Kai Uwe Clausen und Timo Rabe - ihr seid die besten und habt meine Zeit in Kiel zu etwas ganz besonderem gemacht. Zu guter letzt möchte ich mich noch bei meiner Familie bedanken. Besonders bei meiner Mutter und Rolf. Ohne eure riesige und bedingungslose Unterstützung wäre ich nie soweit gekommen. Abschließend möchte ich mich noch bei meiner Verlobten Sina bedanken. Danke für den Rückhalt und den gemeinsamen Alltag. Ich freue mich auf unsere gemeinsame Zukunft! IV List of Publications This thesis consists of three parts. Two of them are based on the following publications, which are published in different journals: 1. J. Junge, S. Froitzheim, T. A. Engesser, J. Krahmer, C. Näther, N. Le Poul and F. Tuczek "Tungsten and Molybdenum Dinitrogen Complex Supported by a Pentadentate Tetrapodal Phosphine Ligand: Comparative Spectroscopic, Electrochemical and Reactivity Studies" Dalton Trans., 2022, 51, 6166-6176. DOI: 10.1039/D1DT04212B 2. S. Froitzheim, J. Junge, C. Barnehl, T. A. Engesser, J. Krahmer, C. Näther and F. Tuczek "Molybdenum tricarbonyl complexes supported by linear PNP ligands: Influence of Pand N-substituents on relative stability, stereoisomerism and on the activation of small molecules" Eur. J. Inorg. Chem. 2023, 26, e202300280. DOI: 10.1002/ejic.202300280 V Kurzzusammenfassung Die synthetische Stickstofffixierung befasst sich mit der Bindung und Derivatisierung des sonst inerten Distickstoffmoleküls mithilfe von Übergangsmetallkomplexen. Im Fokus der Betrachtungen stehen hierbei vor allem Molybdän und Wolfram, die häufig multidentate Liganden binden. Diese Arbeit ist drei in Abschnitte unterteilt. Im ersten Abschnitt ist die Synthese und Charaketrisierung des [W(N 2 )(P Me2 PP Ph2 )] Komplexes beschrieben, welcher auf dem von der Arbeitsgruppe Tuczek entwickelten pentaPod-Ligandsystem basiert. Auch wenn dieser Komplex eine geringere katalytische Aktivität (mit SmI 2 /H 2 O) als sein molybdänbasiertes Analogon hinsichtlich der Erzeugung von Ammoniak zeigte, ist dieser Komplex das erste Beispiel für einen wolframbasierten Distickstoffkomplexes, der überstöchiometrische Mengen NH 3 generieren konnte. Ein Vergleich der Einkristallstrukturen der Mo- & W(N 2 ) Komplexe zeigt nur geringe strukturelle Unterschiede. Um Hinweise auf die Ursache für die unterschiedliche katalytische Aktivität zu finden, wurden elektrochemische und spektroelektrochemische Untersuchungen durchgeführt. Es wurde gefunden, dass die Komplexe große Unterschiede hinsichtlich ihrer Stabilität im oxidierten Zustand aufweisen. Im zweiten Teil dieser Arbeit wird, statt N 2 , CO als Ligand verwendet. Kohlenstoffmonoxid ist isoelektronisch zu Distickstoff, bildet jedoch stabilere Komplexe, was die Untersuchung des Koordinationsverhaltens von Ligandensystemen ermöglicht. Es wurde das Koordinationsverhalten von tridentaten PN Ph P R Liganden (R = Ph 2 , Et 2 , Cyp 2 , i Pr 2 , Cy 2 ) an das Mo(CO) 3 Strukturfragment untersucht. Es wurde gefunden, dass alle Liganden in einer facialen Geometrie koordinierten, obwohl DFT-Rechnungen zeigen, dass eine meridionale Geometrie für sterisch anspruchsvolle Reste bevorzugt wäre. Diese Ergebnisse stehen im Gegensatz zu den in der Literatur bekannten [Mo(CO) 3 PN H P] Komplexen, die für Phosphindonoren mit sterisch anspruchsvollen Substituenten auch eine meridionale Geometrie zeigen. DFT-Studien mit [Mo(CO) 2 PN R P Me ] (R = N, Ph) zeigten unter anderem, dass der Übergangzustand des PN Ph P Me haltigen Komplexes energetisch höher liegt, was auf die Bildung agostischer Wasserstoffbrückenbindungen zurückzuführen ist. Im dritten Teil dieser Arbeit wurden tridentate Ligandsysteme für eine mögliche Fixierung auf einer Goldoberfläche untersucht. Hierzu wurden verschiedene Ansätze gewählt, wobei der Hauptfokus auf der Synthese eines [Mo(CO) 3 PN Ph P Ph ] Komplexes lag, der in para-Position zum N-Donor mit einer Thio-Gruppe versehen war, die als Ankergruppe auf der Oberfläche dienen sollte. Als größte Herausforderung stelle sich hierbei die Einführung der gewählten SCN-Gruppe heraus, da im Zuge dieser Reaktion Br 2 freigesetzt wird, welches die Phosphine oxidierte. Die Schützung der Phosphine mit Borangruppen erlaubte schlussendlich die Einführung der SCN-Gruppe. Leider konnte der gewünschte Ligand dennoch nicht erhalten werden, da Sauerstoffverunreinigungen im Morpholin bei der Entfernung der Borangruppen die Phosphine oxidierten. Nichtsdestotrotz stellt die vorgestelle Route eine vielversprechende Methode zur Darstellung des PN Ph-SCN P Ph Liganden dar, welcher im Anschluss für die Koordination an [Mo(CO)3(cht)] verwendet werden könnte. VI Abstract Synthetic nitrogen fixation deals with the binding and derivatization of the otherwise inert dinitrogen molecule with transition metal complexes. These complexes primarily contain molybdenum and tungsten centers, which mostly bind multidentate ligands. This work is divided into three sections. The first section describes the synthesis and characterization of the [W(N 2 )(P Me2 PP Ph2 )] complex, which is based on the pentaPod ligand system developed by the Tuczek group. Although this complex showed lower catalytic activity (with SmI 2 /H 2 O) than its molybdenum-based analogue in terms of ammonia generation, it is the first example of a tungsten-based dinitrogen complex that is able to generate overstoichiometric amounts of NH 3 . A comparison of the single crystal structures of the Mo- & W(N 2 ) complexes shows only minor structural differences. Electrochemical and spectroelectrochemical investigations were carried out to determine the cause of the different catalytic activities. It was found that the complexes exhibit large differences in terms of their stability in the oxidized state. In the second part of this work, CO is used as a ligand instead of N 2 . Carbon monoxide is isoelectronic to dinitrogen, but tends to form more stable complexes, which allows for the investigation of the coordination behavior of ligand systems. The coordination behavior of tridentate PN Ph P R ligands (R = Ph 2 , Et 2 , Cyp 2 , i Pr 2 , Cy 2 ) to the Mo(CO) 3 fragment was investigated. It was found that all ligands coordinated in a facial geometry, although DFT calculations show that a meridional geometry would be preferred for sterically demanding residues. These results are in contrast to [Mo(CO) 3 PN H P] complexes reported in literature, which do show a meridional geometry for phosphine donors with sterically demanding substituents. DFT studies using [Mo(CO) 2 PN R P Me ] (R = N, Ph) complexes showed, that among other things, the transition state of the PN Ph P Me containing complex is energetically higher, due to the formation of agostic hydrogen bonds. In the third part of this work, tridentate ligand systems were investigated for possible deposition on a gold surface. Various approaches were chosen for this purpose, with the main focus on the synthesis of a [Mo(CO) 3 PN Ph P Ph ] complex, which was modified with a thio group in the para-position to the N-donor, intended to serve as an anchor group to the surface. The introduction of the selected SCN group proved to be the greatest challenge, as Br 2 is released in the course of this reaction, which oxidized the phosphines. Protecting the phosphines with borane groups ultimately allowed the introduction of the SCN group. Unfortunately, the desired ligand could not be obtained because oxygen impurities in the morpholine oxidized the phosphines when attempting to remove the borane groups. Nevertheless, the presented route appears to be a promising method for the preparation of the PN Ph-SCN P Ph ligand, which could subsequently be used for coordination to [Mo(CO)3(cht)]. VII Chapter 2 Scientific Background The amount of energy release even increases with rising temperatures leading to a value of ∆ RH of -109 kJ/mol at 500 °C. [31] Following Le Chateliers principle the equilibrium changes in favor of ammonia by increasing the pressure. In the industrial process usually around 200-300 bar of pressure are applied. [12,30,31] The reduction in volume shifts the reactions equilibrium towards the product side, increasing the yield from 0.13 vol.-% to 17.6 vol.-%. Depending on the contact time between the catalyst and the synthesis gas this yield can vary and yield lower amounts, yet be more economical for a company due to fast reaction times. [30] Both Haber (1919) and Bosch (1931) were awarded the Nobel prize for their accomplishments. In 2007 another Nobel prize was awarded in relation to the Haber-Bosch process. This time it was given to Gerhard Ertl for elucidation of the mechanism on the catalysts surface. [32] The reaction is usually divided into several steps. The first step is the physisorption of dihydrogen and dinitrogen on the iron surface, followed by the dissociative formation of iron nitride species. [30,32,33] The dissociative chemisorption was identified as rate limiting step of the industrial ammonia synthesis as suggested by Emmett. [34] As a consequence of the electronic promotors in the catalyst this reaction is not only exothermic but lowers the very high activation energy that would be associated with a reaction in the gaseous phase. [30,33–36] Interestingly, the dissociation of the dihydrogen molecule seems barrier less and is therefore often neglected in discussions.[37] Fig. 2.1: Energy diagram of the conversion from N 2 and H 2 to NH 3 . The blue reaction pathway describes the reaction on the catalysts surface, the red path the direct reaction in the gaseous phase. Adapted from.[36,37] The dissociative mechanism of the dinitrogen molecule can be separated into three steps. In the first step the dinitrogen molecule is bound in an end-on fashion to the α -iron. Upon transition to a side-on coordination mode the bonds between the nitrogen atoms are weakened. [38,39] Ultimately, this leads to homolytic dissociation of the N ≡ N bonds and the formation of surface nitrides. Although endergonic, chemisorpted hydrogen gets added sequentially. These 4 2.2 Biological Nitrogen Fixation reaction steps aren’t limiting as the hydrogen has a high mobility on the surface and the energetic requirements are easily overcome by the reaction temperature. [30,36,37] After around a century of industrial ammonia synthesis, the Haber-Bosch process evolved to one of the most important chemical processes with ammonia being one of the mostly produced chemicals worldwide. [40] Nonetheless, the process and especially the generation of hydrogen gas are not only very energy-intensive, but also high in expenditure of resources. [22,41] Furthermore, the emission of carbon dioxide is high, especially in view of the challenges posed by climate change. For these reasons, a lot of resources are invested in improving the process with new catalysts [7,42] or finding ways to electrochemically generate ammonia for which the energy needed can be obtained from renewable sources. [22,43,44] The ammonia containing fertilizers are an important factor concerning crop yields and therefore for humanity. Yet nature has found a different way to access and convert atmospheric nitrogen to ammonia. This process is called biological nitrogen fixation. 2.2 Biological Nitrogen Fixation 2.2.1 The Nitrogen Cycle Nitrogenous compounds are essential for plant growth, yet they are unable to produce the needed compounds themselves. Certain microorganism on the other hand are able to convert atmospheric dinitrogen into ammonia. The conversion of dinitrogen to ammonia is the first step in the nitrogen cycle (Figure 2.2), which describes the fixation, conversion and the re-release of nitrogen in nature. [14,45,46] The nitrogen cycle can be separated into three main stages. The first stage is the biological nitrogen fixation (BNF) in which atmospheric dinitrogen is converted into bioavailable compounds by microorganisms. [13,14,46] The fixation and the mechanisms will be discussed in more detail in section 2.2.2. The second stage of the nitrogen cycle is called nitrification. The ammonia generated by nitrogen-fixing bacteria is either assimilated by plants or further oxidized by communities of bacteria generating nitrate. [46,47] Ammonia-oxidizing bacteria and nitrite-oxidizing bacteria convert the ammonia to nitrate via hydroxylamine and nitrite. [48–50] These reactions are an energy source as well as growth factor for the bacteria. [48,49,51] This leads to ammonia and nitrate being the two predominant nitrogen species in the soil.[52] Denitrification, i.e. the degradation of nitrogen species, is the third and final step in the nitrogen cycle. During this step nitrate is reduced to molecular nitrogen and released back into the atmosphere. This is done by sequential reduction of the nitrogen atom. [53,54] The nitrate is reduced to nitrite which in turn is reduced to nitric oxide. The nitric oxide is then converted to nitrous oxide and after a final reduction step dinitrogen is formed. [53,54] Each of these steps is part of an anaerobic respiratory chain and requires its own enzyme as a catalyst. [53–55] A 5 Chapter 2 Scientific Background final process to mention is the dissimilatory nitrate ammonification in which the nitrate is not reduced to N2 but to NH4+ instead. This respiration reaction is not as efficient as the denitrification because the energy gain is less compared to the release of a dinitrogen molecule. Due to this reason the ammonification is mostly used as energy conservation method. [54,56] Not only plants and bacteria are part of the nitrogen cycle. Every living organism takes part in it, yet there is an increasing anthropogenic influence due to extensive fertilization, combustion of fossil fuels, burning of biomass or domestication of animals. [13,45] Due to the success of the Haber-Bosch process and the availability of fertilizers, about half of all nitrogen species in organic matter can be traced back to this process while the other half results from biological nitrogen fixation.[13] Fig. 2.2: Simple illustration of the nitrogen cycle. The three main stages of the nitrogen cycle are the Fixation, Nitrification, and Denitrification. Adapted from[13] 6 2.2 Biological Nitrogen Fixation 2.2.2 Biological Nitrogen Fixation In general the microbial binding and conversion of dinitrogen to ammonia from the air is called biological nitrogen fixation. Similar to the nitrogen cycle itself, the biological nitrogen fixation can be divided into the respective ecosystems in which the associated microbes occur: marine, terrestrial and freshwater systems. [57–59] Depending on the ecosystem different microbes are the main protagonists. Marine nitrogen fixation is mainly associated with cyanobacteria (e.g. Trichodesmium and Richelia). [57,60–62] Terrestrial nitrogen fixation is the second important field of study concerning nitrogen fixation. The most prominent organism being Azotobacter vinelandii. This microorganism was discovered in 1903 and has since been the focus of research related to biological nitrogen fixation. [63] Furthermore, it was used to isolate and investigate the structure of the enzyme enabling its nitrogen fixating capabilities - the nitrogenase (see section 2.3). [64,65] Unlike the marine and terrestrial ecosystems, there have not yet been extensive studies on nitrogen fixation in freshwater and was long considered irrelevant. [59] Due to that reason not much is know, yet some nitrogen fixating species like the freshwater cyanobacterium Anabaena variabilis have been studied. [15,59,66,67] All aforementioned species are diazotrophs meaning they use N2 as nitrogen source in order to grow. In contrast to the Haber-Bosch process, the biological nitrogen fixation is performed under physiological conditions and although a considerable amount of biochemical energy in form of ATP is required (e.g. from photosynthesis in cyanobaterica), the reaction is more efficient in the nitrogenase compared to the industrial heterogeneous catalysis. [66] Besides the ATP as energy source, eight protons and electrons are required for the conversion of dinitrogen to ammonia. The enzymatically catalyzed reaction is as follows:[18] N2+ 8 H++ 8 e−+ 16 MgAT P Nitrogenase −−−−−−−−→ 2NH3+H2+ 16 MgADP + 16 Pi(2.2) 2.2.3 Nitrogenase Nitrogenase was isolated for the first time in 1966 by Bulen and LeComte, but it took another 45 years before the structure could be fully elucidated with the identification of the C 4carbide in the center of the active site. [64,68–70] There are multiple isoforms of the nitrogenase. The most common and best studied is the molybdenum nitrogenase. Due to that reason, this section will focus on the Mo-Nitrogenase only. Alternative forms include a vanadium and an iron nitrogenase. [15–17] These forms are expressed under molybdenum limiting conditions and are much more inefficient compared to the molybdenum containing nitrogenase. [16,17] The nitrogenase complex consists of three protein units: the α2β2 -hetero tetramer called the dinitrogenase (MoFe-protein) and two units of the dinitrogenase reductase (Fe-protein) - one on each end. Two clusters can be found in each αβ -subunit of the nitrogenase. The first one is a Fe8S7 -cluster (P-cluster), which is involved in the electron transfer from the 7 Chapter 2 Scientific Background Fig. 2.3: Ribbon diagram of the Nitrogenase enzyme complex. Located within the MoFe-protein (light blue & purple) are two units of the FeMoco and the P-cluster. On both ends of the MoFe protein binds the dinitrogenase reductase also called Fe-protein (gold & silver). The active components within the Fe-protein are the F-cluster as well as 2 MgATP. This image was created using PDB entry 1N2C and ChimeraX.[71–73] dinitrogenase reductase to the active site of the enzyme. The P-cluster is composed of two Fe4S4 units, but the fourth sulfur atom of each subunit is split, so there are only seven sulfur atoms in the complete cluster (see Figure 2.4B). The reaction shown in eqn. 2.2 takes place at the active center of the dinitrogenase. The active center is an iron-sulfur cluster with an additional molybdenum, hence the name: Iron-Molybdenum-cofactor (FeMoco, see Figure 2.4A). The FeMoco is composed of a molybdenum, seven iron, nine sulfur and a very rare carbidic carbon atom as well as a coordinating homocitrate molecule. Each of the iron atoms is coordinated by three sulfur atoms. Six of the seven iron atoms also bind to the carbidic carbon atom in the center of the cluster, while the seventh iron atom saturates the remaining coordination site with the sulfur of a cysteine. The molybdenum is found in an octahedral geometry, binding three sulfur atoms a homocitrate and completed by a histidine. The cluster was first isolated in 1977, followed by the first structural models in 1978 and the first crystal 8 2.2 Biological Nitrogen Fixation structure 1992. [74–77] The structure revealed by the crystallography showed an unexpected structure of the FeMoco as it appears to feature a hole in the center. [77] Technical progress made it possible to record X-ray spectra with higher resolution, which showed a light atom in this very hole. [78] It took 19 years from the publication of the crystal data to the elucidation of the complete structure with the identification of the carbidic carbon in the center of the cluster. This missing puzzle piece was found by Lancaster et al. and Spatzal et al. and resulted the structure displayed in Figure 2.4A.[74,79,80] Fig. 2.4: Structure of the active site of the Nitrogenase - the iron-molybdenum cofactor (A), the P-cluster with surrounding aminoacids (B) and the F-Cluster (C) of the dinitrogenase reductase. Iron is shown in orange, sulfur in yellow, carbon in gray, nitrogen in blue, oxygen in red and the molybdenum in light blue. This image was created using PDB entries 1N2C, 3U7Q and ChimeraX.[71–73,80] Located between the FeMoco and the Fe-protein is the P-cluster, which is necessary for the electron transfer and the reductive function of the enzyme. [81,82] The electron is transferred to the P-cluster from the [ Fe4S4 ]-cluster (F-cluster, Figure 2.4C) which is located at the interface between the dinitrogenase and dinitrogenase reductase. [82] Within the dinitrogenase reductase the [ Fe4S4 ]-cluster is embedded into the structure via four cysteinyl residues two of which are provided by each γ -chain. [83,84] Within the dinitrogenase reductase two binding sites for MgATP can be found. The MgATP plays a vital role in the mechanism in the substrate reduction activity of the nitrogenase. [85,86] Not only is it a source of energy for the reaction, but the binding results in structural changes within the protein unit that allow the reaction to occur.[81,83] 9 Chapter 2 Scientific Background 2.2.4 Mechanism of Nitrogenase The elucidation of the structure of nitrogenase took several decades, but the structure is not the only aspect that concerns researchers. The mechanism of enzymatic N2 -toNH3 conversion has been the focus of investigation for almost as long and is still not fully understood. The first description of the mechanism was postulated by Thorneley and Lowe after performing a series of kinetic measurements. [87–90] The model unifies assumptions derived from model systems prepared in the context of synthetic nitrogen fixation with kinetic experiments. [87,89,91] Chatt stated that dihydrogen coordinated to a metal center can be displaced by dinitrogen under reductive conditions and that conversion to ammonia occurs through successive protonation and single-electron reduction steps. [89,91] As seen from eqn. 2.2 one equivalent of hydrogen is produced with each turnover of the catalytic cycle. The catalytic cycle described in the Thorneley-Lowe-model is divided into eight distinct one-electron steps (Figure 2.5B). The electrons needed for the catalytic cycle stem from another cycle which is intertwined with the cycle of the FeMoco. The two mechanism are the cycle of the Fe-protein (Figure 2.5A) which transfers the electrons to the FeMoco and the catalytic cycle itself in which ammonia is generated (Figure 2.5B). During the Fe-protein cycle the cluster is found in two oxidation states. The oxidized and the reduced form in which the cluster is found in the oxidation states 1+ and 2+ respectively. [92] Before association and electron transfer to the MoFe-protein two ATP molecules bind to the protein, changing its structural arrangement. [93] This change in the conformation has multiple effects the first being the ability for association of Feand MoFe-Protein. [93–95] In the course of the structural changes, the F-cluster is also rotated closer to the interface between the proteins, bringing it within Van der Waals reach of the P-cluster. [93] As the iron protein is the only known electron source for the reduction and only one electron can be transferred from a protein unit, the dinitrogenase reductase cycle involves association and dissociation after every single-electron transfer. [92,93] Comparing experimental results trying to determine the order of reaction steps in the associated state of the proteins several discrepancies arise. [94] Depending on the method different rate constants were determined. Furthermore, the extremely sensitive protein complex and its environment-dependent properties cause deviations and differences in the experiments, respectively. [94,96–102] Independent of the order of events concerning ATP hydrolysis and electron transfer, phosphate is released after the electron transfer and the protein complex dissociates. The hydrolysis to MgADP reverses the previous conformational change and thus prevents the precondition for the assembly of the Feand MoFe-proteins. [96] 10 2.2 Biological Nitrogen Fixation Fig. 2.5: Illustration of the Thorneley-Lowe-model describing the catalytic cycle of the biological nitrogen fixation. Each step of the cycle is represented by an E with an index reflecting the number of electrons transferred. PDB entries 1N2C [71] and 3U7Q [80] were used for creating the MoFe-, Fe-protein and ferredoxine (PDB 1FRI) images in the iron cycle. Adapted from Einsle and Rees.[89] In the oxidized form the Fe-protein detaches from the MoFe-protein and can be regenerated by reduction. Upon reduction the MgADP is also exchanged with two MgATP molecules, allowing the cycle to start again. For each step, denoted E in Figure 2.5B, the nitrogenase and dinitrogenase reductase must associate and dissociate. [103] Besides the ongoing investigations concerning the order of events as well as the kinetics, determination of the rate limiting step is still part of current discussions in literature.[89,94] The second part of the mechanism is the nitrogen reduction itself. Herein, two parts must 11 Chapter 2 Scientific Background be considered during the cycle in the MoFe protein: The electron transfer via the P-cluster and the reactions at the active site of the enzyme. As described above, the catalytic cycle of the MoFe-protein is divided into eight steps (E 0 -E 7 ) following the Thorneley-Lowemodel. [88–90,104,105] In the first three steps (E 0 -E 3 ) four reduction equivalents are accumulated as bridging hydrides between the iron atoms in the FeMoco. [104–107] The additional two protons are bound to sulfur atoms. [104] The exact structure of the reduced state of the active site is to date not know for certain as the calculated energetics of the cluster show massive variations depending on the DFT method chosen. [104,108,109] The key intermediate of the cycle is found in the E 4 stage which is also called the Janus-Intermediate. In this step, the N2 substrate is bound to the FeMoco. The exact nature of this step is not yet fully understood and different approaches including an active role of the homocitrate ligand as a transitional proton store as well as displacement of the belt sulfurs has been discussed. [110,111] What is known is the fact that the E 4 stage can react in two different directions: back towards the resting state E 0 or continue to protonate and reduce the N2 . [89,105] Continuing the path towards generation of ammonia, two general mechanisms are proposed: a distal and an alternating pathway.[107,112,113] Fig. 2.6: Schematic representation of the alternating and distal mechanism of nitrogenase.[107,113] In the distal path, the terminal nitrogen is first protonated/reduced and then, after the formation of a nitrido species, the second equivalent ammonia is generated. [107,113] This model is also discussed as mechanism in Chatt-type catalysts in synthetic nitrogen fixation (cf. section 2.3.1). The second pathway is the alternating mechanism. Unlike the distal mechanism, in the alternating mechanism, the nitrogen atoms are alternately protonated/reduced and the two equivalents of ammonia are released in the last two steps of the cycle. [107,113] Although 12 2.3 Synthetic Nitrogen Fixation not yet fully proven, the alternate path seems to be more likely. In addition to experimental detection of hydrazine and diazene under catalytic conditions, these compounds can also be used as substrates for the generation of ammonia by nitrogenase.[107,114–116] 2.3 Synthetic Nitrogen Fixation Inspired by nature, synthetic nitrogen fixation deals with the conversion of dinitrogen to ammonia using transition metal complexes. Investigations of the properties of these simpler model systems and their catalytic ability will provide important clues to the mechanism of nitrogenase. Over the decades, however, synthetic nitrogen fixation has developed into a broad branch of research in coordination chemistry, which, in addition to studying different systems and their reactivity as well as developing new catalytic pathways, is also concerned with fundamental questions of coordination chemistry. [19,20,89,117,118] The foundation to synthetic nitrogen fixation was laid by Allen and Senoff (1965) with the synthesis of the first complex bearing a dinitrogen ligand and a ruthenium center ([Ru (N2)(NH3)5 ]X 2 ; X = Br, I, BF 4 , PF 6 ). [119] Addition of NaBH 4 to the complex with iodine counterions also generated the first amounts of ammonia. [119] The first complex coordinating a dinitrogen from nitrogen atmosphere was reported two years later. [120] The coordination of a dinitrogen ligand is observed by infrared spectroscopy as the coordination of the dinitrogen causes a shift to lower wavenumbers compared to free N2 molecule. [117] This shift is commonly referred to as activation. The further the N2 stretch is shifted to smaller wavenumbers, the stronger the dinitrogen ligand is activated with regard to derivatization. [21,121] The cause of the shift is found in the change of the N ≡ N bonds due to the coordination to the metal center. The formation of a bond between the metal center and the nitrogen leads to an electron density transfer, which is described by the Dewar–Chatt–Duncanson model. [122] At first electron density is transferred from the nitrogen to the metal center, known as σ -donation, followed by a π -backdonation, in which electron density is transferred from the metal center back to the ligand (Figure 2.7).[123–125] 13 Chapter 2 Scientific Background 2.3.3 Systems of Peters’ group Although iron is mainly discussed to be the active site of the nitrogenase, synthetic nitrogen fixation mostly evolves around molybdenum and tungsten complexes. The Peters group set another important milestone as they presented the first iron based catalyst capable of N2 -toNH3 reduction, using an iron dinitrogen complex bearing the tris[o-(diisopropylphosphino)phenyl]borane ligand (P3B, Figure 2.12 left).[166] Fig. 2.12: Iron dinitrogen complexes bearing the variations of the ligand for synthetic nitrogen fixation developed by Peters et al.[166,167] Addition of HBAr F and KC8 yielded 7 equivalents of ammonia. [166] The amount of ammonia was increased drastically (to 64 equiv.) by increasing the amount protonand electron sources. [167] Further optimization and additional irradiation of the solution increased the ammonia yield again to 94 equivalents. [168] Unlike the reactions presented by Chatt and Schrock, this catalysis is not carried out under ambient conditions, but at a temperature of -78 °C. Exchanging the reducing agent as well as the acid with less potent compounds ( [Ph2NH2]OTf and Cp2* Co) a yield of 89 equivalents of ammonia were generated - still using the presented iron system. [169] The peculiarity of this combination lies in the transfer mode of the protons and reduction equivalents as the combination of the Cp2* Co with the acid acts as PCET reagent (PCET = Proton Coupled Electron Transfer). [169] This insight was used as a starting point towards electrocatalytic nitrogen fixation, which the group demonstrated in 2018 using [ Cp2* Co] + , [ Fe(N2) ( P3B )] - and anilinium acids in controlled potential electrolysis experiments. In this way, 6.7 equivalents of ammonia could be generated. [170] The main issue of electrocatalytic ammonia generation is the competing hydrogen evolution reaction which is often dominating. [169,170] Early, yet unsuccessful attempts used a mercury pool electrode as the overpotential of mercury allows more negative potentials. [152,171,172] In 2022 the group found an alternative approach towards electrocatalysis by employing a mediator complex that accepts the electrons from the cathode and transfers them as a PCET agent to the actual catalyst (Figure 2.13).[173] 20 2.3 Synthetic Nitrogen Fixation Fig. 2.13: Illustration of the PCET mediated electrochemical N2 -toNH3 reduction by Peters et al., exemplified with a tungsten bisdinitrogen complex. Edited after.[173] A number of catalysts were investigated using this approach, with the best producing 40 equiv. of ammonia. [173] Interestingly, the best catalyst was found to be the very complex employed in the original experiments by Pickett and Talamin in 1985. In the same year the group could also demonstrate direct electrocatalysis on a pyridine based pincer system. Starting from the well known [MoBr3(PNPtBu)] complex 11.7 equiv. NH3 were generated. [174] These impressive results were achieved by very conscious selection of the components involved. Parallel to the efforts invested into the [ Fe(N2) ( P3B )] - complex the Peters group also investigated the influence of the donor trans to the dinitrogen ligand by synthesizing two variants of the P3B ligand. The boron was exchanged by carbon ( P3C ) (Figure 2.12 middle) as well as silicon ( P3Si ) (Figure 2.12 right). [167] However, the catalytic activity of the corresponding iron complexes was lower than that of the [ Fe(N2) ( P3B )] - complex. 47 equiv. of ammonia were found for the [ Fe(N2) ( P3C )] - complex, while [ Fe(N2) ( P3Si )] - generated only 4.4 equiv. [167] Not only variations of the ligands were investigated but also different metal centers. Using the least active P3Si ligand low valent dinitrogen complexes with osmium and ruthenium centers were synthesized. [175,176] Using large amounts of [Ph2NH2][OTf] (1500 equiv.) and Cp2* Co (1800 equiv.) the osmium complex was able to generate 120 equivalents of NH3.[176] 21 Chapter 2 Scientific Background 2.3.4 Systems of the Nishibayashi group The Nishibayashi group focuses on pincer based systems. Per definition pincer ligands are rigid ligands that enforce a meridional coordination geometry. [177] Not only their variability, but also their often high catalytic activity has increased interest in pincer-based systems in recent decades. [178–180] The catalytic activity is often related to the rigidity and high thermal stability of the complexes allowing well defined catalytic reactions. [179–183] Although the term pincer is now used in a broadened sense from its original definition, it usually refers to a tridentate system with 3 donor atoms. Often, amines in the form of pyridines or carbenes are found at the central position. Typical groups for linkage are amines or CH2 -groups but also silicone groups are well known modifications. [182–185] The Nishibayashi group utilizes pincer ligands to develop very active catalysts for synthetic nitrogen fixation. The first system presented by the group capable of catalytic N2 reduction was an end-on bridged binuclear molybdenum complex bearing a PNP pincer with P t Bu 2 donors on each center (Figure 2.14B). The ligand sphere was completed by two more N2 ligands on each molybdenum (Figure 2.14A). [186] The best results were achieved by addition of [LutH][OTf] and Cp 2 Co with 23.2 equiv. of ammonia were generated per dinuclear complex. [186] Based on these initial results, the PNP pincer ligand was modified by variation of the substituents on the phosphine donors and in the para-position of the pyridine. [187,188] Variation of one of the phosphines, and thereby creating asymmetric catalysts, did not increase the amount of ammonia generated, [187] but introduction of different groups on the pyridine ring increased the nitrogen conversion up to 34 equiv. with a methoxy group. [188] This could be improved again (52 equiv. NH3 ) by increasing the added protons and reduction sources. A comparable result with 31 equiv. was found for the 4-methyl-pyridine. [188] In 2015 the system was modified again by exchaning the methoxy group by a ferrocene moiety, which could generate two equivalents more under the same conditions as the methoxy system. [189] Besides the PNP pincer and its modified variants, the group also presented results evolving around a PCP pincer with a N-heterocyclic carbene (NHC) as central donor. [190] Using the [Mo( N2 ) 2 (PCP) 2 ( µ - N2 )] complex and large amounts of [LutH][OTf] and [Cp 2* Co] (1920 and 1440 equiv.) the complex generated 230 equiv. of ammonia.[190] 22 2.3 Synthetic Nitrogen Fixation Fig. 2.14: Selection of complexes used for synthetic nitrogen fixation by Nishibayashi et al. [186,190–192] The dinuclear systems were not the only ones in the scope of the Nishibayashi group. Using the PNP tBu pincer ligand and starting from [MoCl3(thf)3] they generated the nitrido complex [MoNCl(PNP)] (Figure 2.14C), which was also catalytically active towards N2 reduction. Using this complex, [LutH][OTf] and Cp 2 Co a total of 6.6 equiv. of ammonia were generated. [191] This type of nitrido complex was also obtained by use of a tridentate triphosphine ligand, whose catalytic activity was significantly higher in comparison. [192] The ligand is composed of a central phenyl phosphine which is connected to two terminal di-tert-butyl phosphine groups via ethyl bridges (Figure 2.14D). Addition of Cp 2* Co (540 equiv.) and 2,4,6-trimethylpyridinium triflate ([ColH][OTf], 720 equiv.) yielded a total of 63 equiv. of ammonia. [192] Use of these two reagents as proton and electron source, while starting from the molybdenum(III) precursor [MoI3(thf)3 ], in combination with their well established PNP pincer, they managed to obtain a staggering amount of 415 equiv. of ammonia.[193] A major breakthrough was presented that shows parallels to the work of Peters et al. As found for the iron systems, using a PCET agent greatly improved the catalytic activity of the aforementioned pincer systems. The PCET agent used is a combination of samarium diiodide and water or simple alcohols such as methanol, ethylene glycol or ethanol. [194] Using 180 equiv. of SmI2 and ethylene glycol the [MoI3(PNP)] complex bearing two P t Bu 2 groups generated 42.8 equiv. of ammonia while the corresponding nitrido complex generated 50.0 equiv. For the end-on bridged dinitrogen complex bearing the NHC pincer ligand this combination generated 53.3 equiv. of ammonia. [194] The mononuclear trichlorido molybdenum complex bearing this NHC pincer yielded 4350 equiv. of ammonia upon reaction with SmI2 / H2O (14400 equivalents each). This effective catalyst is the first system to use water as proton source to generate ammonia. [194] The use of water proofed to yield larger amounts of ammonia compared to the small alcohols. In 2023 the group presented a trichlorido molybdenum complex bearing a 23 Chapter 2 Scientific Background modified PCP (similar to Figure 2.14 A, with R = CF 3 & H), which is even more active and generated 60.000 ± 4300 equiv. of ammonia per Mo. [195] The difference in reactivity is very well expressed by the use of classical Chatt type complexes which under protic and reductive conditions were not catalytically as they face some problems during the catalytic cycle (cf. section 2.3.1). Application of the SmI2 / H2O allowed for catalytic ammonia generation using these simple molybdenum complexes. [196] Notably, the reaction pathway is expected to differ from the Chatt cycle described above. The trans- [Mo(N2)2(PMe2Ph)4] complex resulted 40 equiv. of ammonia while the in situ generated complex with two equivalents of dpppe (1,5-bis-(diphenylphosphino)-pentane) from MoI3(thf)3 resulted 46 equiv. NH3 . Increasing the amount of PCET agent, this could be scaled to 83 equiv. of ammonia. [196] The approach with samarium diiodide and water/alcohols are the first conditions in which a Chatt-type system is capable of catalytic ammonia generation. The second example for this is presented by our group (section 2.3.5). [197] Interestingly, the tungsten analogues were not catalytic under these conditions. [197] Investigations concerning the mechanism are a major focus of the Nishibayashi group. Several approaches based on experimental findings and DFT calculations a likely mechanism involves a bridged intermediate with a dinuclear molybdenum complex that splits the bridging dinitrogen ligand into nitrides. [197,198] In order to gain further insights into the mechanism, they presented a detailed study on the structure of the SmI2 / H2O structures formed in solution. [199] Other than predicted previously, the experimental results point towards the formation of samarium water and samarium thf complexes, depending on the amount of water, without direct Sm-I bonds.[200] Besides the molybdenum based systems, the group also presented several studies investigating the activity of pincer complexes with other transition metal centers including iron, [201] cobalt, [202] chromium, [203] vanadium, [204,205] rhenium [206,207] and manganese. [208] Although the ligands used all belong to the pincer category and the terminal phosphines chosen bare t Bu 2 substituents, they were very different as were the conditions used in order to obtain catalytic activity. To elaborate on these system a few examples will be explained more detailed. The iron system was based on an anionic PCP pincer with the anionic center being on the central benzene. Upon addition of 8000 equiv. of KC8 and 7360 equiv. of HBAr F 252 equiv. of ammonia and 68 equiv. of hydrazine were generated. [201] Rhenium complexes bearing the pyridine based PNP pincer or a variant with a slight modification in the 4-position generated between 3 and 9 equiv. of ammonia with 800 equiv. of KC8 and 800 equiv. of [HPCy 3 ][BAr F ]. [206] The cobalt system was very different as the ammonia was generated by hydrolysis of a previously formed silylamine. A cobalt dinitrogen complex bearing the PNP pincer was reacted with 600 equiv. KC8 and Me3SiCl . Addition of acid yielded a total of 44 equiv. of ammonia. Scaling this reaction to 6000 equiv. of substrate, 351 equiv. of ammonia were found.[202] The latest approach of the group involves light driven ammonia generation. [200] Using a molybdenum catalysts (e.g. [MoI3(PCP)] ) and photoactive cocatalysts, they were able to 24 2.3 Synthetic Nitrogen Fixation catalytically generate ammonia upon irradiation. [200] The [MoI3(PCP)] complex generated 29.5 equiv. of ammonia. This was increased by slight modification of the ligand. Introduction of a CF3 -group on one side of the benzimidazole backbone increased the yield to 41.3 ± 6.2 equiv. of ammonia.[200] 2.3.5 Systems of the Tuczek group The Tuczek group focuses on Chatt type systems. As described above the classical Chatt systems are facing two major problems (cf. section 2.3.1). The first problem occurs in the nitrido stage, where the oxidation state of the molybdenum center is high, making it a hard Lewis acid that can lead to decoordination of the phosphine donors. The second problem occurs in the regeneration of the catalysts in the last step. In addition to the desired regeneration of the catalyst, a disproportionation reaction occurs that produces inactive dihalide complexes. [21,154] This side reaction leads to a catalyst loss of about 50% with each run of the cycle. [21,154] The Tuczek group is addressing precisely these two points and developing Chatt type systems that attempt to circumvent these problems. This is done in two ways: suppression of the disproportionation reaction by occupation of the trans-position to the dinitrogen, which could also enhance the activation of the ligand, and the decoordination in the high oxidation states is counteracted by increasing the stability by exploiting the chelating effect of multidentate ligands. At first two main strategies were pursued in order to fulfill the set goals. The ligands which meet both criteria for bypassing the problems of the Chatt cycle were characterized by tripodal or tridentate structural motifs. The tripodal ligands are usually based on either an iso-butyl or neo-pentyl backbone. [209–211] The backbone and tripodal geometry of the ligand enforces a facial coordination of the ligand. With addition of a bidentate coligand and reductive conditions mono-dinitrogen complexes were synthesized. [209–211] Besides the iso-butyl and neo-pentyl backbones, our group also succeeded in synthesizing dinitrogen complexes bearing a cyclohexane based tripod and a tripodal system similar to the neo-pentyl system but the quaternary carbon was substituted by silicon. [212–215] Unlike the other tripodal systems, coordination of the cyclohexyl-based ligand requires higher temperatures and reaction times because the donors must be oriented in the axial direction for coordination, which is the less favorable conformation. [212] The silicon substituted ligand was used to investigate the different influence of alkyland aryl-phosphines as donors. It was found that by subsequent exchange of diarylphosphine by dimethylphosphine led to an increase in activation of the N2 ligand. [212] Comparison of the tripod/diphos combination bearing all PMe2 -donors to a mono-dinitrogen complex bearing five PMe3 -donors showed a lower activation for the tripodal complex, illustrating the effect of steric influence on activation. [213,215] To gain further insight into the sterical aspect similar study was performed by Söncksen et al. who chose the iso-butyl (when more than one iPr2 groups were involved) based tripod ligand and sequentially substituted diphenylphosphine by the sterically demanding diisopropylphosphine. Interestingly, 25 Chapter 2 Scientific Background the increase in steric demand lead to the formation of the bis-dinitrogen complexes whereas the complex bearing only arylphosphines formed the mono-dinitrogen complex. [210] A similar study was performed bei Pfeil et al. who prepared ligands containing between 1 and 3 phospholano groups. Phospholano groups are characterized by different behavior as found from regular alkylphosphines. [216–218] For the phospholano tripod - even with three phospholano groups - the mono-dinitrogen complex was found as long as dppm (diphenylphosphinomethane) was used as coligand. [216,218] The activation found for the N2 ligand in the complex with three phospholano groups is lower compared to its PMe2 analogue but higher than the complex with all diphenylphosphine. [216] An important way to tune the activation and thereby the reactivity towards derivatization is careful selection of the trans ligand. [145] The activation should also be increased by substitution of the donor in trans-position to the dinitrogen ligand. Coordination of an amine donor trans to the N2 should increase the activation due to elimination of π -acceptor properties, hence increasing the electron density on the molybdenum and further weaken the N ≡ N-triplebond. [219] Unfortunately, the mono-dinitrogen complex bearing the tripodal ligand with two phosphines and a diethylamine residue could not be obtained, due to the weak coordination behavior of the amine. The product obtained instead was the bisdinitrogen variant.[219] Fig. 2.15: Examples of different nitrogen complexes synthesized by the Tuczek group for synthetic nitrogen fixation.[214,215,218–223] In addition to the tripodal systems, the Tuczek group also investigated nitrogen complexes bearing tridentate and pincer ligands. [145,220,221,224] Following on from the work of George et al., who prepared nitrogen complexes with different tridentate ligands, the group also investigated the influence of different substituents on the donor atoms, as well as the influence of phosphine and amine donors in trans position to the dinitrogen ligand. [145,225,226] The findings of this group regarding the tridentate systems will be discussed in detail in the introduction to chapter 4 as it focuses around tridentate PNP ligands. Marrying the two concepts of tridentate and tripodal ligands, a new singular ligand with pentaphosphine environment emerged, resulting in a pentadentate tetrapodal ligand (pentaPod). 26 2.3 Synthetic Nitrogen Fixation This concept not only eliminates the need for an additional coligand, but also increases the chelate effect even more, stabilizing the corresponding complexes. Furthermore it creates a single site Chatt-type catalyst which would go hand in hand with improved elucidation opportunities regarding the mechanism. The first generation of the pentaPod (P 2Ph PP 2Ph ) featured only PPh2 donors besides the phosphine donor linking the tripodal and tridentate part. Coordination of the ligand to [ MoCl3(thf)3 ] turned out to be quite difficult due to the topology of the precursor and the two possible coordination modes. [222] As the Mo(III) complex is in meridional arrangement, the tridentate part is favored, which was experimentally proven by EPR spectroscopy. Due to this preference the complex has to undergo severe structural changes in order to from the desired mono-dinitrogen complex. It was found that upon reduction several species were formed some of which appear to be coordination polymers with the ligand interconnecting molybdenum centers. [222,227] Adapted from the silicon based tripod ligand, a modified version of the pentaPod with a silicon in the tripodal backbone was synthesized. Using this approach a mono-dinitrogen complex was obtained. Like the tripodal systems the resulting complex was labile against acids and due to the difficult synthesis could not be obtained in its pure form.[222] The problems of the first generation system, which were, among other things, due to the insufficient reactivity differences of the terminal phosphines, were resolved in the second generation of the pentaPod ligand (P 2Me PP 2Ph , Figure 2.16). [222] Substitution of the diphenylphosphine groups on the tripod site by dimethylphosphine introduced nucleophillic gradient into the system with the PMe2 groups, favoring the facial geometry in the Mo(III) stage and enabling the isolation of a mono-dinitrogen complex upon reduction bearing the entire pentaPod ligand. [222,227] The [ Mo(N2) (P 2Me PP 2Ph )] complex not only showed the expected AA’XX’M pattern in the 31 P NMR spectrum but also a very strong activation of the N2 ligand as can be seen from the stretching vibration (νNN = 1929 cm-1) in the IR spectrum.[222,227] Fig. 2.16: The pentaPod concept arose from combination of the tridentate PPP and tripodal PPP ligands.[222,227] Although the activation of the N2 ligand is high and the corresponding hydrazido(2-) complex was obtained, the complex did not show any catalytic behavior under conditions reported by Peters et al. or Nishibayashi et al. [222,223,227,228] The only system found to be active under 27 Chapter 2 Scientific Background these type of conditions found in our group is the PN3P pincer system reported by Stucke et al., which generated 3.12 equiv. of NH3 using Cp2*Cr and [ColH][OTf] (Figure 2.17). [229] Fig. 2.17: [Mo Cl3 (PN 3 P)] complex synthesized by Stucke capable of ammonia generation (3.12 equiv.) upon addition of [ Cp2*Cr ] and [ColH][OTf]. This is the first system of the Tuczek group which exhibited catalytic activity.[229] Although the first attempts towards the catalytic activity of the pentaPod systems were not successful the development continued. Junge et al. succeed in coordinating to the P 2Me PP 2Ph ligand to Rh and Ru. The resulting complexes were used towards activation of small molecules such as CN–or N3–.[230] 28 Chapter 3 Project 1 Even tough a variety of different combinations of acids and reducing agents were tested, no successful catalytic generation of ammonia was achieved until the proton coupled electron transfer (PCET) approach presented by Nishibayashi was applied. Using samarium diiodide and water in thf the pentaPod dinitrogen complex exhibited catalytic behavior. Using this method 25.7 equiv. of ammonia were generated. [231] To investigate the role of the ligand as an aspect of the newly discovered catalytic activity, similar mono-dinitrogen complexes with pentaphosphine environment were synthesized and tested towards their catalytic activity. The complexes prepared were a neo-pentyl based tripodal PPP ligand bearing all terminal PPh2 groups with a dmpm (dimethylphosphinomethane) coligand and the analogue tridentate variant with the prPPHP ligand, which is also used as tridentate building block for the pentaPod ligand. Both complexes were tested toward their catalytic activity under identical conditions as the pentaPod complex, however, only substochiometric amounts of ammonia were found, proofing the effectiveness of the pentaPod concept. [231] The system is a single coordination site Chatt-type catalyst and one of the main intermediates of the Chatt cycle is the hydrazido(2-) complex, Engesser et al. were also able to isolate and thoroughly characterize the corresponding NNH2 complex after protonation with HBAr F . [231] In order to prove the role of the [ Mo(NNH2) (P 2Me PP 2Ph )]BAr F complex in the catalytic cycle, this complex was also used as catalyst with SmI2 / H2O . Due to stability issues in the required solvent, the complex was generated in situ and added to the PCET-agent containing solution. This approach yielded 26.1 equiv. of ammonia proving the hydrazido(2-) stage to be an entry point in the catalytic cycle.[231] Chatt systems do not only involve molybdenum but also tungsten complexes. In fact the first systems of the Chatt-type complexes that yielded the best results for ammonia generation (90 %) were the cis-[ W(N2)2(PMePh2)4 ] complex by Chatt et al. and the [W(NNH2)(OTs)(dppe)2]+ complex employed for electrocatalysis by Pickett and Talamarin. [151–153] Besides molybdenum, tungsten is another transition metal often found in synthetic nitrogen fixation. Due to its important role and the effectiveness of the pentaPod concept, combination of tungsten and the pentaPod complex is an interesting approach, which will be the first project presented in this thesis. 29 Chapter 3 Project 1 Aiming at determining the role of the metal in the described redox processes, we carried out the same experiments for [Mo(N 2 )(P Me2 PP Ph2 )] (1). CV studies revealed several differences compared to the W–N 2 complex. For instance, the first oxidation process at E 1/2 (1) was found to be not fully reversible at low scan rate (v< 0.1 V s −1 ). Moreover, a supplementary irreversible anodic peak at E pa (3) (ca. −0.7 V vs. Fc + /Fc) was detected for v< 0.5 V s −1 (Fig. 5A). Further oxidation at E pa (2) was also accompanied by a supplementary oxidation peak at E pa (4) which disappeared at high scan rate (Fig. 5B). At last and conversely to complex 2, the Mo complex displayed no sign of reversibility for the second reduction process at E pa (2) for high values of v(Fig. 5B and Fig. S13B†). Spectroelectrochemical studies performed with the Mo–N 2 complex 1also revealed some differences to its W congener 2. While oxidation at E pa (1) led to disappearance of the N 2 stretching band at 1942 cm −1 , no new IR-detectable species could be detected in the 1800–2100 cm −1 frequency range (Fig. 5C). Moreover, upon back reduction, the initial spectrum could only be partially recovered. Interestingly, analysis at lower frequencies (1400 to 1700 cm −1 ) displayed further spectral changes upon electrochemical oxidation (Fig. 5D). Pushing the potential until E pa (2) induced even larger modification of the spectra in this frequency region. Returning to the initial potential value partially restored the N 2 stretching band at 1942 cm −1 (Fig. 5D). Notably, the changes in the 1400 to 1700 cm −1 frequency range were not observed for the tungsten complex upon oxidation at E pa (2) (Fig. S13D†). Altogether, these electrochemical and spectroelectrochemical investigations essentially revealed two results. First, they Scheme 3 Proposed mechanism for the oxidation processes of the tungsten (A) and molybdenum (B) pentaPod N 2 complexes. Fig. 5 CVs (E/V vs. Fc + /Fc) at a Pt working electrode (diam. 1 mm) of [Mo(N 2 )(P Me2 PP Ph2 )] (0.4 mM) in THF/NaBPh 4 0.02 M (A) at v= 0.02 V s −1 and (B) for v= 0.02 V s −1 (black), 0.05 V s −1 (red), 0.1 V s −1 (green), 0.2 V s −1 (blue), 0.5 V s −1 (cyan), 1 V s −1 (pink) and 2 V s −1 (olive); the numbers (1), (2), (3) and (4) on the graphics are related to the redox systems, see details in text. (C) Infrared spectra of [Mo(N 2 )(P Me2 PP Ph2 )] (15 mM) in THF/NaBPh 4 20 mM recorded during in situ spectroelectrochemical measurements before (black) and after oxidation at E pa (1) (red), then returning back to the initial potential (blue); (D) the same as (C) except that oxidation at E pa (1) (red) is followed by oxidation at E pa (3) (green), E pa (2) (blue), then returning back to the E pa (1) (cyan) and finally the initial potential (magenta). Paper Dalton Transactions 6170 |Dalton Trans., 2022, 51,6166–6176 This journal is © The Royal Society of Chemistry 2022 Published on 11 March 2022. Downloaded by Christian Albrechts Universitat zu Kiel on 1/20/2023 8:50:42 AM. View Article Online 36 3.1 Tungsten and molybdenum dinitrogen complexes supported by a pentadentate tetrapodal phosphine ligand: comparative spectroscopic, electrochemical and reactivity studies clearly show that one-electron oxidation of the neutral W and Mo dinitrogen complexes occurs at almost the same E 1/2 (1) value (−1.16 V and −1.13 V vs. Fc + /Fc, respectively). This result is reasonable on the basis of experimental data obtained from NMR and IR spectroscopies as well as X-ray diffraction (vide infra), which have shown minor differences between the two neutral Mo and W complexes in solution and in the solid state. It is also in line with electrochemical investigations reported on Mo and W bis-N 2 complexes supported by P Ph N Me P Ph ligands, exhibiting a difference of only 20 mV for the first oxidation potential. 24 Hence, our results suggest that the structural properties of the W and Mo mono-oxidized dinitrogen species are very similar. The second information which can be taken from electrochemistry and spectroelectrochemistry is that, although W I (N 2 ) + and Mo I (N 2 ) + pentaPod complexes can be reduced at similar potential values, they display very different stabilities in THF (cf. Scheme 3): whereas the mono-oxidized tungsten N 2 species seems to be highly stable, yielding back the neutral N 2 complex upon reduction, spectroelectrochemistry suggests that its Mo analogue evolves rapidly (sec) towards a new Mo I species, probably a Mo I (thf) + complex. Likewise, the W II (N 2 ) 2+ species appears at the second oxidation process as fairly unstable (msec) and may exchange its N 2 ligand by a THF ligand. Upon back reduction, N 2 re-binds to the metal center since the N 2 stretching band is detected. In the case of the molybdenum complex, oxidation beyond E pa (1) leads to a more complicated situation and probably induces N 2 –THF ligand exchange. The anodic peak at E pa (3) may be thus ascribed to the oxidation of Mo I (thf) + species. At last, comparison with redox data reported for other dinitrogen Mo and W complexes offers interesting information (see Table 2). In particular, oxidation potentials of mononuclear trans-bis-N 2 W and Mo complexes are clearly more positive than those of W and Mo pentaPod complexes by 150–370 mV, likely resulting from replacement of one dinitrogen ligand by a P-donor. In some cases, such as for trans-[Mo(N 2 ) 2 (dppe) 2 ], the CV reversibility for the first oxidation was found to be dependent on the experimental conditions. Whereas Chatt et al. described a reversible system in THF/NBu 4 BF 4 for the first oxidation process, 26 Elson reported an irreversible anodic peak at room temperature when using a THF/MeOH (26% v/v) mixture with LiCl or LiClO 4 as supporting electrolyte. 23 In the latter case, a trans-[Mo I (dppe) 2 (MeOH) 2 ] + species was characterized resulting from the release of N 2 according to a dissociative pathway. On the other hand, for trans-[Mo(N 2 ) 2 (depe) 2 ], 25 longer timescale yielded NuN bond cleavage and Mo IV nitride formation, differently to the complex 1. Reactivity of Mo and W complexes towards acids and SmI 2 /H 2 O In case of the molybdenum pentaPod complex [Mo(N 2 ) (P Me2 PP Ph2 )] (1) the corresponding hydrazido(2-) complex could be generated by addition of Brookhart’s acid (HBAr F ,[H (OEt 2 ) 2 ][BAr F ], 2.5 equiv.). Notably, this complex also exhibited catalytic activity towards ammonia generation, indicating that it is an intermediate of the catalytic cycle. 16 Treatment of the tungsten dinitrogen complex 2with 3 equiv. of HBAr F was similarly found to generate the corresponding hydrazido(2-) complex 3-BAr F (Scheme 4; for IR data see ESI Fig. S15 and Table S7†). Using smaller amounts of acid did not lead to a pure product. The 31 P NMR spectrum of 3-BAr F exhibits an AA′MXX′ pattern (Fig. 6, see Fig. S16–S26†for full NMR data), demonstrating retention of the pentaphosphine environment (a signal at about 0 ppm shows a small impurity which most likely is caused by protonation of the ligand, also tiny amounts of free ligand can be observed in the enlargement in Fig. 7). However, all signals of 3-BAr F have undergone a high field shift compared to 2(Fig. 7), indicating an increased shielding of the P-donors in the former complex which is attributed to an elongation of the metal–P bonds and a loss of σ-donation to the metal centre. The particularly large high-field shift of the M-signal upon protonation of the N 2 -complex is due to the large trans-influence of the π-donating hydrazido(2-) ligand, weakening the W–P ax bond. This analysis is supported by DFT calculations which show an increase of the W–P bond lengths upon going from 2to 3-BAr F , with a larger elongation of the Scheme 4 Protonation of [W(N 2 )(P Me2 PP Ph2 )] (2) to [W(NNH 2 ) (P Me2 PP Ph2 )]X 2 (X = BAr F , Al(pftb) 4 )(3). Table 2 Electrochemical data for [LM I/0 ] dinitrogen species Complex E 1/2 /V vs. Fc + /Fc Conditions Ref. [W(N 2 )(P Me2 PP Ph2 )] (2)−1.16 THF/NaBPh 4 This work [Mo(N 2 )(P Me2 PP Ph2 )] (1)−1.13 THF/NaBPh 4 This work trans-[Mo(N 2 ) 2 (depe) 2 ]−1.01 THF/Pyr 4 FAP 25 −0.99 a THF/NBu 4 BF 4 26 trans-[W(N 2 ) 2 (dppe) 2 ]−0.82 THF/NBu 4 [B(C 6 F 5 ) 4 ]24 trans-[W(N 2 ) 2 (dppe)(dppp)] −0.79 THF/NBu 4 [B(C 6 F 5 ) 4 ]24 a Experimentally measured at E 1/2 =−0.43 V vs. SCE. Dalton Transactions Paper This journal is © The Royal Society of Chemistry 2022 Dalton Trans., 2022, 51,6166–6176 | 6171 Published on 11 March 2022. Downloaded by Christian Albrechts Universitat zu Kiel on 1/20/2023 8:50:42 AM. View Article Online 37 Chapter 3 Project 1 W–P ax bond length in comparison to W–P eq (Table S6†). Notably, protonation of 2with [H(OEt 2 ) 2 ][Al(OC(CF 3 ) 3 ) 4 ] (HAl (pftb) 4 ) led to similar results (cf. Fig. S27†). The 31 P– 31 P coupling constants of the hydrazido(2-) complex 3-BAr F show some differences compared to the parent dinitrogen complex 2(Table 1). While the cis couplings among the equatorial phosphines are almost unchanged (ΔJ=1–2Hz),the corresponding trans coupling gets smaller by about 30 Hz. On the other hand, the cis couplings of the equatorial phosphines with the axial P donor are increased by 17 and 13 Hz, respectively. Similar protonation-induced changes have been observed in the corresponding molybdenum system (Table 1). 15,16 Comparison of the hydrazido(2-) complexes of tungsten and molybdenum show similar differences as for the dinitrogen complexes, except that the changes in the cis coupling constants between the equatorial phosphines and the axial P-donor are smaller than for the corresponding dinitrogen complexes (Table 1). The 1 H– 15 N, 15 N– 15 N and 15 N– 31 P coupling constants show nearly identical, the corresponding values for the tungsten complexes always being slightly smaller than for their molybdenum counterparts (see Table S8 in ESI†). Nevertheless, the 1 H– 15 N coupling constant of 3-BAr F (92.0 Hz) is still larger than for a W-hydrazido(2-) complex supported by two bidentate ligands (80 Hz). 27 Analogous to the dinitrogen complexes, DFT calculations indicate that the bond lengths and bond angles for the molybdenum and tungsten hydrazido(2-) complexes are very similar (Table S6†). Moreover, the calculated N–N bond lengths (∼1.31 Å) correspond to those obtained for classic Chatt-type hydrazido(2-) complexes. 28 In order to investigate a potential catalytic activity of the new tungsten complex, a 0.1 M solution of samarium diiodide and water (5 ml) was treated with a THF solution of 2(2 µmol in 1 ml THF) under nitrogen in a 50 mL Schlenk flask at room temperature. After the solution turned yellow the amount of produced ammonia was determined by the indophenol method. 29 In contrast to its molybdenum congener 1,complex 2just generated a slightly over-stoichiometric amount of ammonia (2.75 ± 0.23 eq., 5% yield related to the reducing) agent. Whereas ammonia could be stoichiometrically produced by protonation of [W(N 2 ) 2 (PMePh 2 ) 4 ], 11 a lack of (chemo)catalytic NH 3 formation from N 2 is well known for tungsten systems (see above). 13,14 Thus, 2is the first W complex that is able to generate more than 2 equivalents of NH 3 in relation to the metal centre upon addition of protons and reductant. However, it appears that just a small fraction of the dinitrogen complex is regenerated, suggesting that most of the catalyst is converted to adifferent complex that is inactive towards nitrogen reduction. In this context we note that addition of water to the tungsten dinitrogen complex without samarium diiodide resulted in the decomposition of the complex. Summary and conclusions The tungsten dinitrogen complex [W(N 2 )(P Me2 PP Ph2 )] (2)supported by a pentadentate tetrapodal phosphine ligand (P Me2 PP Ph2 ) has been synthesized and characterized regarding its electronic structure and reactivity, allowing comparison with the analogous molybdenum complex [Mo(N 2 )(P Me2 PP Ph2 )] (1). Reaction of [W(N 2 )(P Me2 PP Ph2 )] (2) with samarium iodide/water was found to mediate a slightly overstoichiometric formation (2.75 ± 0.23 eq.) of NH 3 from N 2 which makes 2the first tungsten complex generating more than 2 equivalents of ammonia from N 2 . Notably, the analogous molybdenum complex 1catalytically generates 25.7 eq. NH 3 from N 2 . 16 This contrasting behaviour of tungsten vs. analogous molybdenum complexes regarding the catalytic conversion of N 2 to NH 3 is well known in the literature, 8,13 but not fully understood. With a ν NN value of 1901 (2)vs. 1929 cm −1 (1) the activation of the N 2 ligand is higher in the W complex than in its Mo analogue, which therefore cannot be the reason for the differing behaviour. Apart from that, however, differences in structural parameters are small. Correspondingly, a single-crystal structure determination Fig. 6 Experimental and simulated 31 P{ 1 H} NMR spectra of 15 N 2 -3-BAr F (a) and 3-BAr F (b). (c) Overall spectrum of 3-BAr F in diethylether-d 10 . Fig. 7 Comparison of the 31 P{ 1 H} NMR spectra of 2(black) and 3-BAr F (red). Paper Dalton Transactions 6172 |Dalton Trans., 2022, 51,6166–6176 This journal is © The Royal Society of Chemistry 2022 Published on 11 March 2022. Downloaded by Christian Albrechts Universitat zu Kiel on 1/20/2023 8:50:42 AM. View Article Online 38 3.1 Tungsten and molybdenum dinitrogen complexes supported by a pentadentate tetrapodal phosphine ligand: comparative spectroscopic, electrochemical and reactivity studies of 2only showed minor differences of bond distances and angles with respect to 1. Likewise, the 31 Pand 15 N-NMR-spectroscopic properties of 2were found to be quite similar to those of 1, apart from characteristic differences in chemical shifts and coupling constants. Based on the strong activation of N 2 ,2 could be converted to the hydrazido(2-) derivative 3by protonation with HBAr F , similar to 1, rendering comparison to the analogous Mo–NNH 2 complex possible as well. In order to obtain information about the electronic–structural properties of the tungsten and molybdenum pentaPod systems, electroand spectroelectrochemical investigations were performed on 1and 2. Both dinitrogen complexes exhibit remarkably similar redox potentials. Moreover, two main systems can be detected in oxidation. For tungsten, the first system is found to be reversible and the second, being irreversible at low scan rates, gets more reversible at higher scan rates. By contrast, the first system is only partially reversible for the molybdenum complex and irreversible for the second system. The origin of these differences could be elucidated with spectroelectrochemistry: upon oxidation of 2, a stable W I N 2+ complex is formed whereas for the molybdenum complex 1 loss of N 2 occurs, probably going along with a ligand exchange and formation of a THF-bound complex. For W, loss of N 2 only occurs upon further oxidation of the W I N 2+ complex. In conclusion, the spectroelectrochemical studies as well as the voltammetric studies showed that both systems can regenerate the zerovalent dinitrogen complexes around the same potential. Furthermore, a stable tungsten dinitrogen complex is already formed at the level of a W(I) intermediate, exhibiting a fairly activated N 2 ligand with a ν NN of 1951 cm −1 in solution. By contrast, the corresponding Mo(I) dinitrogen complex is thermally unstable. A theoretical mechanism evaluated for N 2 - reduction of the molybdenum pentaPod system indicated that PCET to the Mo(N 2 )-complex, generating a diazenido(-) intermediate, is energetically more favourable for a Mo(I)- than for a Mo(0)-species. 16 Based on the results presented here, the tungsten complex 2would fit much better to such a scenario than its molybdenum analogue 1. Nevertheless, NH 3 formation mediated by 2is only slightly overstoichiometric. The reason(s) for this observation must therefore lie in some other stage of the catalytic cycle. Further investigation of this question is underway. Experimental section All syntheses were performed under N 2 or argon atmosphere using standard Schlenk line and glovebox techniques. The solvents were dried and freshly distilled under argon prior to use. [H(OEt 2 ) 2 ][Al(OC(CF 3 ) 3 ) 4 ] was received from the working group of I. Krossing in Freiburg i. Br.. All other reagents were commercially available and were used as received. [WCl 4 (PMePh 2 ) 2 ], 18 [Mo(N 2 )(P Me2 PP Ph2 )] (1) 15 and [Mo(NNH 2 ) (P Me2 PP Ph2 )] 16 were prepared according to the literature. NMR spectra were recorded with a Bruker AVANCE III HD Pulse Fourier transform spectrometer operating at frequencies of 400.13 MHz ( 1 H), 376.50 ( 19 F), 161.98 MHz ( 31 P), 128.38 ( 11 B), and 40.56 MHz ( 15 N). Referencing was performed either using the solvent residue signal (5.32 ppm for CD 2 Cl 2 , 3.58 ppm for thf-d 8 and 7.16 ppm for C 6 D 6 ) or TMS (δ 1 H = 0 ppm), CFCl 3 (δ 19 F = 0 ppm), 85% H 3 PO 4 (δ 31 P = 0 ppm), BF 3 ((δ 11 B= 0 ppm), and CH 3 NO 2 (δ 15 N = 0 ppm) serving as substitutive standards. IR spectra were recorded at RT on a Bruker Vertex70 FT-IR spectrometer using a broadband spectral range extension VERTEX FM for full mid and far IR in the range of 6.000–80 cm −1 . Electrochemical studies were performed in a glovebox (Jacomex) (O 2 < 1 ppm, H 2 O < 1 ppm) with a home-made 3-electrode cell (WE: Pt or glassy carbon, RE: Pt wire in a 1 mM Fc + /Fc, 0.02 M THF/NaBPh 4 solution, CE: Pt). Ferrocene was added at the end of the experiments to determine the exact redox potential values. The potential of the cell was controlled by an AUTOLAB PGSTAT 100 (Metrohm) potentiostat monitored by the NOVA© software (Metrohm). The working electrodes were polished over a 1 μm alumina slurry with water, sonicated in H 2 O (18.2 Ωcm) and acetone, then dried with N 2 flush. Thin layer IR spectroelectrochemistry was carried out with a previously described set-up consisting of a commercial IR Si ATR probe (Artphotonics), which can fit into a thin space created on the surface of a glassy carbon electrode. 30 The thin layer between the probe and the electrode allows fast (seconds) electrolysis, hence time-resolved monitoring of the electrochemical reaction. Detection of the IR signal (2 cm −1 resolution, one spectrum every 10 s) was obtained by using a FTIR optic-fiber-coupled spectrometer purchased from Arcoptix (FTMIR-FC-120-LN2). Single crystal structure determination Data collection was performed using an Imaging Plate Diffraction System (IPDS-2) from Stoe & Cie with Mo-Kα radiation. A numerical absorption correction was performed using X-Red and X-Shape of the software package X-Area. The structures were solved with SHELXT 31 and structure refinement was performed against F 2 using SHELXL-2018 32 The C–H hydrogen atoms were positioned with idealized geometry (methyl H atoms allowed to rotate but not to tip) and were refined isotropic with U iso (H) = 1.2U eq (C) (1.5 for methyl H atoms) using a riding model. CCDC 2127859 (2) contains the supplementary crystallographic data for this paper.† Computational details Calculations of the tungsten dinitrogen [W(N 2 )(P Me2 PP Ph2 )] (2) and hydrazido(2-) complexes [W(NNH 2 )(P Me2 PP Ph2 )] 2+ (3) were conducted with the ORCA 4.2.1 program package. 33 Geometries were optimized on PBE0 34 /def2-TZVP 35 level of theory. In addition, Grimme’s dispersion correction with Becke–Johnson damping (D3BJ) 36 and the RI approximation using the def2-TZVP/J fitting basis set 37 were used. Dalton Transactions Paper This journal is © The Royal Society of Chemistry 2022 Dalton Trans., 2022, 51,6166–6176 | 6173 Published on 11 March 2022. Downloaded by Christian Albrechts Universitat zu Kiel on 1/20/2023 8:50:42 AM. View Article Online 39 Chapter 3 Project 1 [WCl 3 (κ 3 -P Me2 PP Ph2 )] 1.11 g (1.53 mmol) of [WCl 4 (PMePh 2 ) 2 ] and 1.24 g (1.83 mmol) of P Me2 PP Ph2 were dissolved in 50 ml toluene and stirred for 3 h at 70 °C. The mixture was filtrated and concentrated in vacuo to 5 ml and 10 ml of diethyl ether was added. The green precipitate was filtered and washed with 10 ml diethyl ether and 10 ml n-pentane. Drying in vacuo gave a dark green solid. Yield: 1.26 g (1.30 mmol, 85%). Anal. calcd for C 39 H 53 Cl 3 P 5 W: C, 48.4; H, 5.52; found: C, 48.9; H, 5.65. IR (300 K): v ˜= 3067 (w), 3051 (w), 2954 (sh), 2919 (m), 2867 (sh), 2801 (w), 1589 (m), 1568 (w), 1551 (vw), 1480 (m), 1455 (vw), 1430 (m), 1412 (w), 1381 (w), 1330 (vw), 1297 (w), 1279 (w), 1260 (vw), 1241 (w), 1182 (w), 1156 (w), 1119 (w), 1096 (m), 1069 (w), 1025 (w), 1000 (w), 945 (m), 917 (m), 880 (m), 843 (w), 805 (vw), 739 (s), 693(s), 639 (vw), 617 (w), 577 (vw), 543 (w), 508 (s), 479 (m), 442 (vw), 425 (w), 404 (vw), 362 (vw), 340 (sh), 323 (sh), 294 (s), 271 (vs), 252 (sh), 221 (vw), 208 (vw), 178 (w), 158 (w), 133 (m), 120 (w) cm −1 . Raman (300 K): v ˜= 6054 (m), 2982 (w), 2913 (s), 1586 (s), 1455 (vw), 1414 (w), 1306 (vw), 1210 (vw), 1186 (w), 1160 (w), 1100 (w), 1073 (vw), 1028 (m), 1000 (vs), 960 (w), 786 (w), 688 (w), 619 (w), 400 (vw), 328 (w), 258 (w), 150 (vw) cm −1 . [W(N 2 )(P Me2 PP Ph2 )] (2) 290 mg (300 µmol) of [WCl 3 (κ 3 -P Me2 PP Ph2 )] were dissolved in 20 ml THF and added to sodium amalgam prepared of 2 ml Hg and 200 mg (8.7 mmol) of sodium. The reaction mixture was stirred for 16 h under an atmosphere of nitrogen. The supernatant red solution was transferred into another flask and the solvent was removed in vacuo. The residue was resolved in 10 ml of diethyl ether and filtered over neutral alumina. The orange solution was concentrated to dryness and shed with small amounts of cold n-pentane or n-hexane. Drying in vacuo gave an orange solid. Crystals suitable for X-ray single crystal diffraction were obtained by slow diffusion of n-pentane into a benzene-d 6 solution of 2. Yield: 118 mg (133 µmol, 44%). Anal. calcd for C 39 H 53 N 2 P 5 W: C, 52.7; H, 6.01; N, 3.15. Found: C, 53.4; H, 6.04; N, 2.30. The nitrogen value is too low because of the thermal instability of the product. 1 H NMR (400.13 MHz, d 6 -benzene, 300 K): δ= 0.51 (d, 2 J= 5.2 Hz, 6H, PMe 2 ), 0.81 (d, 2 J= 7.7 Hz, 2H, PCH 2 ), 0.97 (m, 7H, Me 2 PCH 2 ,CH 3 ), 1.27 (m, 4H, PCH 2 ), 1.93 (m, 4H, CH 2 CH 2 CH 2 ), 2.03 (m, 2H, Ph 2 PCH 2 ), 2.43 (m, 2H, Ph 2 PCH 2 ), 6.41 (m, 5H, PPh 2 ), 6.75 (m, 2H, PPh 2 ), 6.86 (m, 6H, PPh 2 ), 7.09 (m, 4H, PPh 2 ), 7.16 (m, 1H, PPh 2 ), 7.28 (m, 2H, PPh 2 ) ppm. 13 C NMR (100.61 MHz, d 6 -benzene, 300 K): δ= 16.3 (m, 2C, PMe 2 ), 23.9 (m, 4C, CH 2 CH 2 CH 2 ), 26.2 (m, 2C, PMe 2 ), 30.3 (s, 2C, Ph 2 PCH 2 ), 34.6 (m, 1C, PCH 2 ), 38.6 (m, 2C, PCH 2 ), 40.6 (s, 1C, H 3 C), 42.0 (m, 1C, H 3 CC), 43.0 (m, 2C, Me 2 PC), 126.4 (s, 8C, PPh 2 ), 127.3 (d, 2 J= 7.4 Hz, 4C, PPh 2 ), 127.8 (s, 2C, PPh 2 ), 128.1 (m, 4C, PPh 2 ), 129.1 (m, 2C, PPh 2 ), 147.2 (m, 2C, PPh 2 ), 150.0 (m, 2C, PPh 2 ) ppm. 31 P{ 1 H} NMR (161.98 MHz, d 6 -benzene, 300 K): δ=−4.8 (m, 2 J(P AX′ ,P A′X ) = 85.7, 2 J(P AM ,P A′M ) = 11.2, 2 J (P AX ,P A′X′ )=−11.1, 2 J(P A ,P A′ ) = 17.4 Hz, 2P, PPh 2 ,P A /P A′ ), −21.3 (tt, 2 J(P MX ,P MX′ ) = 18.5 Hz, 1P, P M ), −35.8 (m, 2 J(P X ,P X′ ) = 4.5 Hz, 2P, PMe 2 ,P X /P X′ ) ppm. IR (300 K): v ˜= 3139 (vw), 3069 (w), 3049 (w), 2994 (vw), 2959 (m), 2920 (sh), 2906 (m), 2855 (m), 2812 (sh), 1982 (w), 1901 (vs, ν(N–N)), 1805 (vw), 1750 (vw), 1585 (w), 1571 (w), 1481 (m), 1448 (w), 1431 (s), 1418 (sh), 1373 (w), 1328 (vw), 1293 (w), 1258 (s), 1218 (vw), 1184 (vw), 1153 (w), 1089 (s), 1072 (m), 1048 (m), 1025 (s), 970 (w), 925 (m), 905 (s), 875 (m), 832 (w), 798 (s), 737 (s), 693 (vs), 662 (s), 625 (s), 612 (s), 544 (vw), 506 (vs), 475 (m), 429 (w, ν(W–N)), 402 (s), 372 (vw), 345 (m), 313 (w), 290 (w), 263 (vw), 240 (m), 191 (vw) cm −1 . Raman (300 K): v ˜= 3053 (s), 2963 (w), 2912 (vs), 2888 (s), 2863 (sh), 1903 (s, ν(N–N)), 1587 (s), 1572 (w), 1451 (vw), 1434 (vw), 1418 (vw), 1405 (vw), 1297 (vw), 1274 (vw), 1186 (w), 1156 (w), 1093 (m), 1029 (m), 1001 (vs), 933 (vw), 910 (vw), 670 (w), 637 (vw), 618 (w), 505 (vw), 431 (w, ν(W–N)), 403 (vw), 364 (vw), 350 (vw), 315 (vw), 292 (vw), 257 (vw), 242 (w), 193 (w) cm −1 . 15 N-[W(N 2 )(P Me2 PP Ph2 )] ( 15 N-2) was prepared in a similar fashion as 2under a 15 N 2 atmosphere and characterized by IR, Raman, 31 Pand 15 N NMR spectroscopy. 15 N{ 1 H} NMR (40.56 MHz, d 10 -Et 2 O, 300 K): δ=−18.7 (d, 1 J NN = 7.5 Hz, 1N, N β ), −48.4 (m, 1N, N α ) ppm. 31 P{ 1 H} NMR (161.98 MHz, d 6 - benzene, H 3 PO 4 , 300 K): δ=−4.8 (m, 2 J(P AA′ N α ) = 2.7 Hz, 3 J (P AA′ N β ) = 0.3 Hz, 2P, PPh 2 ,P A /P A′ ), −21.3 (ttdd, 2 J(P M ,N α )= 13.9, 3 J(P M N β ) = 1.6 Hz, 1P, P M ), −35.8(m, 2 J(P XX′ N α ) = 2.6 Hz, 3 J(P AA′ N β ) = 0.7 Hz, 2P, PMe 2 ,P X /P X′ ) ppm. IR (300 K): v ˜= 1840 (vs, ν(N–N)), 421 (w, ν(W–N)) cm −1 . Raman (300 K): v ˜= 1840 (s, ν(N–N)), 419 (w, ν(W–N)) cm −1 . [W(NNH 2 )(P Me2 PP Ph2 )] (BAr F ) 2 (3-BAr F ) A portion of 99.0 mg (97.9 µmol) of [H(OEt 2 ) 2 ][BAr F ] was dissolved in 0.4 mL of diethyl ether and added to 29 mg (32.6 µmol) of 3 in 0.6 mL of diethyl ether. After stirring for 5 min at room temperature the solvent was removed in vacuo, affording a light brown solid of nominal composition (determined by integration of the 1 H NMR signals) [W(NNH 2 ) (P Me2 PP Ph2 )](BAr F ) 2 ·3HBAr F . The high fluorine content precluded an elemental analysis. 1 H NMR (400.13 MHz, d 6 - benzene, 300 K): δ= 0.85 (d, 2 J= 7.8 Hz, 6H, PMe 2 ), 1.03 (m, 3H, CH 3 ), 1.59 (d, 2 J= 9.7 Hz, 2H, PCH 2 ), 1.73 (m, 8H, Me 2 PCH 2 ,PCH 2 ), 2.24 (m, 4H, CH 2 CH 2 CH 2 ), 2.66 (m, 2H, Ph 2 PCH 2 ), 2.98 (m, 2H, Ph 2 PCH 2 ), 6.75 (m, 6H, PPh 2 ), 6.90 (m, 2H, PPh 2 ), 7.35 (m, 6H, PPh 2 ), 7.42 (s, 22H, BAr F ), 7.62 (m, 38H, BAr F ), 7.68 (m, 4H, PPh 2 ), 7.77 (m, 2H, PPh 2 ) ppm. 13 C NMR (100.61 MHz, d 6 -benzene, 300 K): δ= 16.7 (m, 2C, PMe 2 ), 19.3 (s, 4C, CH 2 CH 2 CH 2 ), 20.8 (m, 2C, PMe 2 ), 26.0 (m, 2C, PCH 2 ), 28.5 (s, 2C, PCH 2 ), 35.2 (m, 1C, PCH 2 ), 35.4 (s, 1C, H 3 CC), 36.0 (m, 2C, Me 2 PC), 116.0 (s, 4C, CH(BAr F )), 123.4 (q, 1 J= 271.7 Hz, CF 3 ), 127.4 (s, 8C, CH(BAr F )), 128.0 (m, 8C, CCF 3 ), 128.0 (m, 8C, PPh 2 ), 128.5 (s, 2C, PPh 2 ), 129.6 (d, 2 J= 15.0 Hz, 4C, PPh 2 ), 130.7 (m, 2C, PPh 2 ), 131.4 (d, 2 J= 11.6 Hz, 4C, PPh 2 ), 136.1 (d, 1 J= 44.1 Hz, 2C, PPh 2 ), 136.1 (d, 1 J= 44.1 H 2C, PPh 2 ), 137.0 136.1 (d, 1 J= 45.8 H 2C, PPh 2 ), 160.2 (m, 4C, BC) ppm. 11 B NMR (128.38 MHz, d 6 -benzene, 300 K): δ=−6.97 (s, 2B, BAr F ) ppm. 19 F NMR (376.50 MHz, d 6 -benzene, 300 K): δ =−63.4 (s, 48F, BAr F ) ppm. 31 P{ 1 H} NMR 161.98 MHz, d 10 - Et 2 O, 300 (K): δ=−15.9 (m, 2 J(P AX′ ,P A′X ) = 53.9, 2 J(P AM ,P A′M )= 28.2, 2 J(P AX ,P A′X′ )=−12.2, 2 J(P A ,P A′ ) = 16.1 Hz, 2P, PPh 2 ,P A / Paper Dalton Transactions 6174 |Dalton Trans., 2022, 51,6166–6176 This journal is © The Royal Society of Chemistry 2022 Published on 11 March 2022. Downloaded by Christian Albrechts Universitat zu Kiel on 1/20/2023 8:50:42 AM. View Article Online 40 3.1 Tungsten and molybdenum dinitrogen complexes supported by a pentadentate tetrapodal phosphine ligand: comparative spectroscopic, electrochemical and reactivity studies P A′ ), −45.8 (m, 2 J(P MX ,P MX′ ) = 31.3, 2 J(P X ,P X′ ) = 6.3 Hz, 2P, PMe 2 ,P X /P X′ ), −48.6 (tt, 1P, P M ) ppm. IR (300 K): v ˜= 3312 (w, ν(N–H)), 3076 (vw), 2960 (w), 2924 (m), 2853 (w), 1790 (vw), 1611 (m), 1443 (w), 1352 (s), 1272 (vs), 1159 (m), 1114 (vs), 1092 (sh), 998 (w), 934 (w), 918 (vw), 886 (s), 838 (s), 809 (m), 743 (m), 711 (s). 698 (w), 680 (s), 669 (s), 617 (vw), 609 (vw), 580 (w), 568 (vw), 516 (w), 507 (w), 488 (w), 449 (m), 403 (w), 392 (w), 380 (vw), 366 (m), 353 (sh), 331 (vw), 283 (w), 259 (w), 246 (vw), 210 (vw), 189 (vw), 151 (vw) cm −1 . Raman (300 K): v ˜= 3074 (w), 3023 (vw), 2960 (8w), 2932 (m), 2906 (m), 2882 (sh), 2852 (w), 1611 (m), 1593 (s), 1523 (vw), 1464 (w), 1364 (s), 1107 (w), 1030 (w), 1003 (vs), 952 (w), 939 (vw), 914 (vw), 840 (vw), 801 (s), 745 (w), 704 (s), 689 (w), 675 (w), 617 (vw), 287 (w), 262 (w), 244 (vw), 235 (w), 183 (w), 159 (m) cm −1 . 15 N-[W(NNH 2 )(P Me2Ph2 )](BAr F ) 2 ( 15 N-3-BAr F ) was prepared in a similar fashion starting from 15 N-2 and characterized by IR, Raman, 31 Pand 15 N NMR spectroscopy. 15 N{ 1 H} NMR (40.56 MHz, d 10 -Et 2 O, 300 K): δ=−48.4 (m, 1N, N α ), −237 (m, 1N, N β ) ppm. 31 P{ 1 H} NMR 161.98 MHz, d 10 -Et 2 O, 300 (K): δ= −15.9 (m, 2 J(P AA′ N α ) = 5.0 Hz, 2P, PPh 2 ,P A /P A′ ), −45.8 (m, 2 J (P XX′ N α ) = 5.5 Hz, 2P, PMe 2 ,P X /P X′ ), −48.6 (ttdd, 2 J(P M ,N α )= 20.4 Hz, 3 J(P M N β ) = 6.6 Hz, 1P, PM) ppm. IR (300 K): v ˜= 3308 (w, ν(N–H)), 554 (w, ν(W–N)) cm −1 . [W(NNH 2 )(P Me2 PP Ph2 )][Al(pftb) 4 ] 2 (3-Al(pftb) 4 ) was prepared similar to 3-BAr F with [H(OEt 2 ) 2 ][Al(OC(CF 3 ) 3 ) 4 ] as acid. Conflicts of interest The authors declare no conflict of interest. Acknowledgements The authors gratefully acknowledge the SEA-EU program supported by Agence Nationale de la recherche (ANR-19GURE-0001), PHC Procope Campus France (46652ZL) and Programm des Projektbezogenen Personenaustauschs Frankreich (PROCOPE) 2021–2023 (Deutscher Akademischer Austauschdienst, Projekt-Kennziffer 57560918) for funding. The authors thank Prof. Ingo Krossing for providing us with [H(OEt 2 ) 2 ][Al(OC(CF 3 ) 3 ) 4 ]. References 1(a) Y. Tanabe and Y. Nishibayashi, Chem. Soc. Rev., 2021, 50, 5201; (b) M. J. Chalkley, M. W. Drover and J. C. Peters, Chem. Rev., 2020, 120, 5582; (c) N. Stucke, B. M. Flöser, T. Weyrich and F. Tuczek, Eur. J. Inorg. Chem., 2018, 2018, 1337. 2(a) L. C. Seefeldt, Z.-Y. Yang, D. A. Lukoyanov, D. F. Harris, D. R. Dean, S. Raugei and B. M. Hoffman, Chem. Rev., 2020, 120, 5082; (b) C. van Stappen, L. Decamps, G. E. Cutsail, R. Bjornsson, J. T. Henthorn, J. A. Birrell and S. DeBeer, Chem. Rev., 2020, 120, 5005; (c) A. J. Jasniewski, C. C. Lee, M. W. Ribbe and Y. Hu, Chem. Rev., 2020, 120, 5107. 3 R. Schlögl, Angew. Chem., Int. Ed., 2003, 42, 2004. 4(a) A. D. Allen and C. V. Senoff,Chem. Commun., 1965, 24, 621; (b) M. D. Fryzuk, Chem. Commun., 2013, 49, 4866. 5(a) J. Chatt, A. J. Pearman and R. L. Richards, Nature, 1975, 253, 39; (b) M. Hidai, K. Tominari, Y. Uchida and A. Misono, J. Chem. Soc. D, 1969, 1392; (c) M. Hidai, K. Tominari and Y. Uchida, J. Am. Chem. Soc., 1972, 94, 110. 6 C. J. Pickett and J. Talarmin, Nature, 1985, 317, 652. 7 D. V. Yandulov and R. R. Schrock, Science, 2003, 301, 76. 8 R. R. Schrock, Acc. Chem. Res., 2005, 38, 955. 9(a) K. Arashiba, Y. Miyake and Y. Nishibayashi, Nat. Chem., 2011, 3, 120; (b) A. Eizawa, K. Arashiba, H. Tanaka, S. Kuriyama, Y. Matsuo, K. Nakajima, K. Yoshizawa and Y. Nishibayashi, Nat. Commun., 2017, 8, 14874. 10 Y. Ashida, K. Arashiba, K. Nakajima and Y. Nishibayashi, Nature, 2019, 568, 536. 11 J. Chatt, A. J. Pearman and R. L. Richards, J. Chem. Soc., Dalton Trans., 1977, 1852. 12 P. Garrido-Barros, J. Derosa, M. J. Chalkley and J. C. Peters, ChemRxiv, 2021, DOI: 10.33774/chemrxiv-2021-j95jg. 13 K. Arashiba, K. Sasaki, S. Kuriyama, Y. Miyake, H. Nakanishi and Y. Nishibayashi, Organometallics, 2012, 31, 2035. 14 D. V. Yandulov and R. R. Schrock, Can. J. Chem., 2005, 83, 341. 15 S. Hinrichsen, A. Kindjajev, S. Adomeit, J. Krahmer, C. Näther and F. Tuczek, Inorg. Chem., 2016, 55, 8712. 16 T. A. Engesser, A. Kindjajev, J. Junge, J. Krahmer and F. Tuczek, Chem. –Eur. J., 2020, 26, 14807. 17 E. Carmona, A. Galindo, M. L. Poveda and R. D. Rogers, Inorg. Chem., 1985, 24, 4033. 18 Inorganic syntheses, ed. F. Basolo and R. J. Angelici, Wiley, New York, 1990, vol. 28. 19 S. L. Apps, A. J. P. White, P. W. Miller and N. J. Long, Dalton Trans., 2018, 47, 11386. 20 F. B. Ogilvie, J. M. Jenkins and J. G. Verkade, J. Am. Chem. Soc., 1970, 92, 1916. 21 (a) A. Galindo, E. Gutierrez, A. Monge, M. Paneque, A. Pastor, P. J. Perez, R. D. Rogers and E. Carmona, J. Chem. Soc., Dalton Trans., 1995, 3801; (b) E. Carmona, J. M. Marin, M. L. Poveda, J. L. Atwood and R. D. Rogers, J. Am. Chem. Soc., 1983, 105, 3014. 22 J.-M. Savéant, Elements of Molecular and Biomolecular Electrochemistry: An Electrochemical Approach to Electron Transfer Chemistry, Wiley, Hoboken NJ, 2006. 23 C. M. Elson, Inorg. Chim. Acta, 1976, 18, 209. 24 C. J. Weiss, A. N. Groves, M. T. Mock, W. G. Dougherty, W. S. Kassel, M. L. Helm, D. L. DuBois and R. M. Bullock, Dalton Trans., 2012, 41, 4517. 25 A. Katayama, T. Ohta, Y. Wasada-Tsutsui, T. Inomata, T. Ozawa, T. Ogura and H. Masuda, Angew. Chem., Int. Ed., 2019, 58, 11279. Dalton Transactions Paper This journal is © The Royal Society of Chemistry 2022 Dalton Trans., 2022, 51,6166–6176 | 6175 Published on 11 March 2022. Downloaded by Christian Albrechts Universitat zu Kiel on 1/20/2023 8:50:42 AM. View Article Online 41 Chapter 3 Project 1 26 J. Chatt, H. Wasif, G. J. Leigh, H. Neukomm, C. J. Pickett and D. A. Rankin, J. Chem. Soc., Chem. Commun., 1980, 1024. 27 C. J. Weiss, J. D. Egbert, S. Chen, M. L. Helm, R. M. Bullock and M. T. Mock, Organometallics, 2014, 33, 2189. 28 K. H. Horn, N. Böres, N. Lehnert, K. Mersmann, C. Näther, G. Peters and F. Tuczek, Inorg. Chem., 2005, 44, 3016. 29 (a) M. W. Weatherburn, Anal. Chem., 1967, 39, 971; (b) M. P. E. Berthelot, Rep. Chim. Appl., 1859, 1, 284. 30 C. Garcia Bellido, L. Álvarez-Miguel, D. Miguel, N. Lalaoui, N. Cabon, F. Gloaguen and N. Le Poul, ChemElectroChem, 2021, 8, 1899. 31 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Adv., 2015, A71,3. 32 G. M. Sheldrick, Acta Crystallogr., Sect. C: Struct. Chem., 2015, C71,3. 33 F. Neese, Wiley Interdiscip. Rev.: Comput. Mol. Sci., 2018, 8,33. 34 J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3686. 35 (a) A. Schäfer, C. Huber and R. Ahlrichs, J. Chem. Phys., 1994, 100, 5829; (b) A. Schäfer, H. Horn and R. Ahlrichs, J. Chem. Phys., 1992, 97, 2571; (c) F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005, 7, 3297. 36 (a) S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys., 2010, 132, 154104; (b) S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem., 2011, 32, 1456. 37 (a) K. Eichkorn, O. Treutler, H. Öhm, M. Häser and R. Ahlrichs, Chem. Phys. Lett., 1995, 240, 283; (b) K. Eichkorn, F. Weigend, O. Treutler and R. Ahlrichs, Theor. Chem. Acc., 1997, 97, 119; (c) F. Neese, J. Comput. Chem., 2003, 24, 1740. Paper Dalton Transactions 6176 |Dalton Trans., 2022, 51,6166–6176 This journal is © The Royal Society of Chemistry 2022 Published on 11 March 2022. Downloaded by Christian Albrechts Universitat zu Kiel on 1/20/2023 8:50:42 AM. View Article Online 42 3.2 Electrochemical and Electrocatalytic Studies 3.2 Electrochemical and Electrocatalytic Studies In order to investigate the dinitrogen complexes, which are very sensitive to air and moisture, the electrochemical measurements need to be performed under an inert atmosphere. Besides cyclic voltammetry and coulometry, investigations regarding the electrocatalytic activity of the complexes are in the scope of the present thesis. Since at the time, utilization of the glove box was not yet an option, appropriate vessels for the desired applications had to be developed. The requirements for measurements and catalytic studies are different, yet the cells should be similar. Glass vessels for electrochemistry were already available, which leaves the requirement for an air tight lid. For this, a lid composed of two elements was designed. The lower part (Figure 3.1 black) features a groove which fits a glass lip of the vessels fits. The cell is sealed by a second element made from PTFE with a groove that holds an O-ring to assure an air tight fit. The two parts of the lid are locked and held together by four screws. It is very important that the screws are not tightened too much as the PTFE of the lid could crack upon too much stress. The PTFE lid of the cell has multiple holes, which provide the possibility for different connections. Each hole perfectly fits a small rubber septum with a tight fit. In the intended uses, the rubber septa serve as insertion points for quasi-reference, auxiliary electrodes and for the addition of preparations, e.g. by a syringe pump. Fig. 3.1: Inert gas cell designed for electrochemical analytics and use in electrocatalytical experiments In the center of the upper lid is a hole that is larger than the other ones. This central hole can be used in two ways. For electrochemical measurements such as cv, a working electrode can be inserted which however requires an additional fitting. By cutting off the top of a small rubber septum and fixing it to the electrode with parafilm allows for an air tight fit. Using this method cyclic voltammetry measurements under inert conditions were achieved. The gas tightness of the cell was determined by means of gas chromatography. The capability of the 43 Chapter 3 Project 1 system was tested by measuring the [W(N2)(P2MePP2Ph)] complex (Figure 3.3). Fig. 3.2: The cyclic voltammograms of the [ W(N2) (P 2Me PP 2Ph )] complex (c = 0.06 mmol/L) measured with the cells build for the electrochemical investigations (thf, 0.1 mol/L TBAPF 6 ). Display is the overview voltammogram (left) and a measurement of the redox events of the complex (right). Two redox systems are associated with the complex. The first reversible redox system can be assigned to the oxidation/reduction of W 0⇄ W I and the second, quasi-reversible system to the WI⇄WII. The cyclic voltammograms obtained for this complex and its molybdenum analogue are identical to the ones obtained in cooperation with Le Poul from the Université de Bretagne Occidentale Brest. This demonstrates the functionality of the system. The cyclic voltammograms shown in Figure 3.3 are referenced to the potentials known from the measurements in Brest. The main issue at the time was that the complexes were measured against a quasi-reference electrode and no ferrocene/ferrocenium was added to compensate for the shift of the Ag-quasi-reference. Nonetheless, the cvs demonstrate the functionality of the cell design. The second application of the cell is meant for electrocatalytic investigations. By inserting a glass tube, containing a G4 frit at the bottom, into the larger central hole, the working electrode is separated from the auxiliary electrode allowing for electrolysis experiments. The central hole is fitted with an additional o-ring inside a groove in the middle of the PTFE lid, sealing the cell. Due to the interaction of the o-rings between the two parts of the lid and in the center hole with the solvent used, the cell is sealed even better. The polymer of the o-rings takes up some of the solvent vapor, leading to an increase in their volume. As for the cyclic voltammetry configuration, the air thightness of the cell in electrocatalytic configuration was verified by GC measurements of the cell atmosphere after different time periods and after experiments. Neither water nor oxygen could not be detected in any of the performed measurements. 44 3.2 Electrochemical and Electrocatalytic Studies Fig. 3.3: Setup for investigations towards electrocatalytic activity of the complex using a mercury pool electrode. In order to perform investigations regarding the electrocatalytic activity of the complexes, a mercury pool electrode was used. This type of electrodes has two main advantages. The first advantage is the big surface of the electrode, eliminating any diffusion limitations, film formation or deposition on the surface by utilization of a stirring bar. The second advantage is the overpotential of the mercury electrode. As protons are required for ammonia generation, application of low potentials would lead to the hydrogen evolution reaction (HER) at the electrodes surface as the dominant reaction, even before ammonia would be generated. Mercury has a very high overpotential, allowing relatively low potentials to be applied without hydrogen evolution. It is important to note that the vessel used for a mercury pool cathode differs from the ones used for measurements in the way that the connector for the working electrode is located on the bottom of the vessel. The main disadvantage of the cell shown in Figure 3.1 is the volume required for measurements. Due to that reason an identical lid system for vessels with a smaller volume was build. One of the vessels was modified for electrocatalytic studies by inserting a thick platinum wire into the bottom of glass. 45 Chapter 4 Project 2 PN3P pincer and a tripodal triphos ligand via appended 4ethinylphenyl units to TATA (triazatriangulenium) platforms. The resulting functionalized tricarbonyl complexes were deposited on gold and studied with a range of surface-spectroscopic methods.[10,22] As alternative headgroups for surface studies, molybdenum dinitrogen (or carbonyl) complexes supported by linear PNP ligands may be envisioned.[25] Notably, introducing an ethinyl group in 4-position of the central phenyl ring of the PNPhP ligand (cf. Scheme 1, bottom right) would enable similar coupling strategies to platforms or organic surface layers as applied earlier for PN3P and triphos ligands and derived complexes (see above).[10,22] The molybdenum tricarbonyl complex [Mo(CO)3(PNPhPPh)] supported by the PNPhPPh ligand (Scheme 1, lower right) has been prepared by Keskin et al. and investigated regarding to his structural and spectroscopic properties.[26] Various electronic factors were explored to explain the preferred formation of the fac isomer.[26] In order to increase the activation of small molecules such as CO or N2in molybdenum complexes supported by the PNPhPPh ligand,[25] its terminal phenyl groups have to be replaced by more electron-donating alkyl residues. Correspondingly we synthesized a series of linear PNPhPRligands with different alkyl residues R=Me, Et, Pln, Cyp, iPr, Cy, tBu and studied the structural and electronic properties of derived molybdenum tricarbonyl complexes, using X-ray structure analysis and spectroscopy. As expected, all new complexes exhibit a shift of the CO-stretching frequencies to lower wavenumbers with respect to the parent complex. Surprisingly, however, all of these complexes are found in the fac geometry, regardless of the steric demand of the alkyl substituents. This is in strong contrast to analogous complexes supported by PNHP ligands, where both, mer and fac isomers can be identified. In an effort to understand this result (and provide arguments going beyond those advanced by Keskin et al. to account for the fac constitution of the parent [Mo(CO)3(PNPhPPh)] complex; see above), hypothetical mer-fac isomerization processes are modeled by DFT for both PNPhP and PNHP complexes. These calculations indicate that mer-fac isomerization is kinetically hindered in Mo-tricarbonyl complexes supported by PNPhP ligands whereas it is thermally allowed in the analogous PNHP systems. Results and Discussion Synthesis and X-ray Structure Determination Most of the PNP ligands (1a–1 f and 1h) were prepared by reaction of N,N-bis(2-chloroethyl)aniline with a secondary phosphine in the presence of n-butyllithium (Scheme 2). For the PNPhPiPr ligand (terminal isopropyl groups, phenyl substituent on the central nitrogen; 1 f), which was previously reported by Curley et al.,[19,27] N,N-bis(2-tosylethyl)aniline instead of the chlorinated analog was employed. The latter route was also used by Kostas for the synthesis of PNPhPPh (1a).[6] All ligands could be obtained in pure form and were reacted with the precursor [Mo(CO)3(cht)] in toluene (cht=cycloheptatriene), leading to [Mo(CO)3(PNPhPR)] complexes with R=Ph (2a), Et (2c), Cyp (2e), iPr (2f), and Cy (2 g; cf. Scheme 3 and experimental section, general procedures A and B). As expected, the solid tricarbonyl complexes showed sufficient air-stability for characterization via infrared spectroscopy. Notably, not all target compounds could be prepared in this fashion. Specifically, the Mo tricarbonyl complexes supported by the ligands PNPhPMe (2b), PNPhPln (2d) and PNPhPtBu (2h) could neither be isolated in pure form nor characterized in solution. For 2b and 2h, products were obtained which were insoluble. IR investigations indicate that most likely mixtures of different complexes are formed. On the other hand, DFT calculations suggest that the complex 2h containing tert-butyl groups may not form due to steric reasons (Table S2, Table S3, Figure S63). In particular, longer bond lengths are found in the calculated structure of fac-[Mo(CO)3(PNPhPtBu)] (2h) as compared to [Mo- (CO)3(PNPhPiPr)] (2f), which might impede a complete coordination of the ligand. Factors influencing the relative stabilities of the target complexes in solution will be analyzed later in this study, employing DFT calculations (see below). For complex [Mo(CO)3(PNPhPEt)] (2c), a single-crystal X-ray structure determination could be performed, showing a facial coordination mode of the PNP ligand (Figure 1, Table 1). This is analogous to the related complex fac-[Mo(CO)3(PNPhPPh)] (2a) reported by Keskin et al.[26] The biggest structural difference between the two fac- [Mo(CO)3(PNP)] complexes 2a and 2 c is exhibited by the MoN bond length which is 0.0703 Å shorter in 2 c than in 2a. On the Scheme 2. Synthesis scheme of the PNP ligands used in this study (Pln=phospholano, PC4H8; Cyp=cyclopentyl). Scheme 3. Synthesis of [Mo(CO)3(PNPhPR)] complexes (PR=P). Wiley VCH Montag, 28.08.2023 2399 / 317727 [S. 2/15] 1 Eur. J. Inorg. Chem. 2023, e202300280 (2 of 14) © 2023 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202300280 10990682c, 0, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202300280 by Cochrane Germany, Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 52 4.1 Molybdenum tricarbonyl complexes supported by linear PNP ligands: Influence of Pand N-substituents on structure, stability and on the activation of small molecules other hand, the MoP distances are increased in 2 c relative to 2a by 0.018 and 0.041 Å. This is due to the fact that compared to the diphenylphosphine groups of 2a, the diethyl phosphine groups of 2c have a larger sigma donor strength and a smaller capacity for π-backbonding. Interestingly, the increased MoP bond length found in 2c appears to entail a shorter MoN bond, suggesting a leveling effect of the central amine group on the overall binding properties of the PNP ligand. This is also evident from the fact that the MoC distances are very similar in 2a and 2c. Nevertheless, vibrational spectroscopy reveals that the CO ligands of 2c are stronger activated than in 2a (cf. next section). Infrared and Raman Spectroscopy Detailed information on the bonding and activation of CO ligands is provided by IR and Raman spectroscopy. Moreover, the CO-stretching vibrations also reflect the mer/fac isomerism of the title complexes. For an octahedral complex with facial M(CO)3coordination, group theory predicts two bands in the CO stretching region of the IRand in the Raman-spectrum (A1 and E). Due to the actual Cs-symmetry of complexes 2 a,2c,2 e, 2f, and 2 g (Figure 4 and Figure S36, S44, S48, and S52), E splits into A’’ and A’(2), in addition to A’(1) which corresponds to the totally symmetric, in-phase stretching mode. In the Raman spectrum, the A’(1) band is found to be very weak for the facial isomer, while the A’’ and A’(2) bands are strong, clearly visible in case of 2c (Figure 2). The meridional isomer, on the other hand, would exhibit an intense A’(1) band.[28] Therefore, vibrational spectroscopy shows that our [Mo(CO)3(PNPhPR)] complexes always adopt the facial coordination mode in the solid state. Notably, the infrared and Raman spectra of the parent complex 2a recorded by us differ from those published by Keskin et al.[26] While the A’(1) and A’’ vibrations appear at comparable frequencies, the frequency of the A’(2) mode given by these authors (1782–1796 cm1) is considerably lower than observed by us (1808/1804 cm1; cf. Table 2). Moreover, the intensity of the A’(1) peak in the Raman spectrum is fairly high, which would in principle speak against the assignment of a fac geometry to 2a, as inferred by these authors on the basis of the vibrational spectra. We do not know the reason for this discrepancy, but note that our IRand Raman data for 2a are consistent with those obtained for the other complexes. Interestingly, for all investigated complexes except 2g, four instead of two bands are observed for the A’’ and A’(2) stretching vibrations in the Raman spectra (Table 2). Complex 2c, in which this splitting is most pronounced, shows a difference of Δν =12 and 10 cm1between the A’(2) and A’’ bands, respectively. This splitting is also observed when spectra of single crystals of 2c Figure 1. Molecular structure of fac-[Mo(CO)3(PNPhPEt)] (2c). The hydrogen atoms have been omitted for clarity. Thermal ellipsoids are shown at 50% probability. Table 1. MoX bond lengths [Å] in complexes 2a and 2 c. MoX bond[a] [Mo(CO)3(PNPhPEt)][b] (2c) [Mo(CO)3(PNPhPPh)][c] (2a) Δ MoN1 2.442 2.512 0.070 MoP1 2.514 2.496 0.018 MoP2 2.507 2.466 0.041 MoC41(ax) 1.938 1.933 0.005 MoC21(eq) 1.978 1.977 0.001 MoC31(eq) 1.973 1.979 0.006 [a] Numbering according to Figure 1; [b] this study; [c] Ref. [26]. Figure 2. IR and Raman spectrum [cm1] of [Mo(CO)3(PNPhPEt)] (2c). The CO stretching vibrations are assigned based on a Cssymmetric Mo(CO)3 fragment. Table 2. Experimental Raman frequencies [cm1] of the CO stretching vibrations in [Mo(CO)3(PNPhPR)] complexes (R=Ph (2a), Et (2c), Cyp (2 e), iPr (2f), Cy (2g)). Complex A’(1) A’’[a] A’(2)[a] [Mo(CO)3(PNPhPPh)] (2a) 1921 1829/1819 1808/1804 [Mo(CO)3(PNPhPEt)] (2c) 1906 1820/1810 1786/1774 [Mo(CO)3(PNPhPCyp)] (2e) 1907 1812/1795 1785 [Mo(CO)3(PNPhPiPr)] (2g) 1901 1802/1794 1788/1780 [Mo(CO)3(PNPhPCy)] (2f) 1906 1810 1791 [a] In case of splitting both bands are listed. Wiley VCH Montag, 28.08.2023 2399 / 317727 [S. 3/15] 1 Eur. J. Inorg. Chem. 2023, e202300280 (3 of 14) © 2023 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202300280 10990682c, 0, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202300280 by Cochrane Germany, Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 53 Chapter 4 Project 2 were measured, so it does not originate from impurities or a mixture of different compounds. In the crystal structures of 2c and 2a, more than one molecule exists within the unit cell (Z= 8 and 4, respectively.)[26] In this case, the vibrations of individual molecules in the unit cell may couple (for Z=2, e.g., into inphase and out-of-phase combinations), and the corresponding bands or peaks split (Figure 3). This effect (factor group or Davydov splitting) has already been observed for molybdenum carbonyl complexes.[29] In addition to the coordination mode, information about the activation of the carbonyl ligands can be derived from the frequencies of the stretching vibrations of the Mo(CO)3fragment. This is influenced by the σ-donation and π-acceptance properties of the phosphine and amine donors. From the literature, it is known that the difference in σ-donation properties of tertiary phosphines with different residues is small.[30] In contrast, their π-acceptor properties change significantly, which then influences the vibrational frequency of the CO in trans position. Correspondingly, for the complexes bearing different alkyl phosphines (2c,2e,2f, and 2 g) the observed A’(1) CO stretching frequencies are fairly similar (Table 2), the biggest difference (Δ(A’(1))=6 cm1) being found between the systems with PCyp2and PiPr2substituents. In contrast, the A’(1) frequency of complex 2a bearing phenyl phosphines (1921 cm1) is 14 cm1higher than 2e, the alkyl phosphine complex with the highest frequency. This conforms to aryl phosphines being more π-backbonding than their alkyl counterparts. Overall, the activation of the CO ligands as a function of the substituents on the phosphines is found to increase in the following sequence: PPh2<PEt2�PCyp2�PCy2<PiPr2 NMR Spectroscopy 31P and 13C NMR spectroscopy in solution provides information on the mer-fac isomerism of complexes 2 a,2c &2e–2g in solution via the 31P resonances of the P-donors and the 13C atoms of the carbonyl ligands coupling with the phosphines, respectively. Importantly, the low-field shifts observed for the 31P-signals of the complexes (ca. 15–22 ppm) are characteristic for a facial binding geometry. In meridional complexes, larger low-field shifts of the phosphine signals are observed.[31] Whereas the fraction of fac complexes exceeds 99%, traces of mer isomers can also be detected (<1%; cf. Figure 4 and Table S1). Other byproducts appear to be tetracarbonyl complexes which result from decomposition of their tricarbonyl analogs, as reported in the literature.[26] Complete NMR data are given in the Supporting Information. Complementary information can be derived from the 13CNMR spectra. Due to the low natural abundance of this carbon isotope (<1.1%), molecules containing more than one 13C nucleus can be neglected.[32] Notably, a 13C nucleus next to a 31P atom leads to different chemical environments for otherwise equivalent 31P nuclei. Correspondingly, the coordination mode can be derived from the signal patterns emerging in the 13C NMR spectrum by coupling of the 31P nuclei with the 13C atoms statistically found on the different carbon positions. Specifically, ameridional coordination geometry is reflected by two triplets, one for the CO trans to the amine group and one for the CO ligands trans to each other. In contrast, a facial geometry gives rise to a triplet for the axial carbonyl carbon (resulting from the equal cis couplings to the equatorial phosphines) and, for the equatorial carbon atoms, to the X part of an ABX spin system with A and B represented by two equatorial 31P atoms (A=PA, B=PBand X=13C; see Figure 3 for complex 2c as an example).[22,28] In agreement with the results from 31P-NMR spectroscopy (see above), the facial isomers are found with ratios of more than 99% for all complexes (2 a,2c &2 e–2g). For these species, the X part of the ABX spectrum (see above) corresponds to a six-line signal (see Figure 5, top, “b” and bottom scheme). The distance between lines 2 and 5 is equal to jJAX +JBX j, while the separation between lines 1 and 6 (lines 3 and 4) equals 2jD++ D-j(2jD+-D-j, respectively). D+and D-are defined by Equation (1). Furthermore, the relative intensities of the inner Figure 3. Splitting of the carbonyl stretches of [Mo(CO)3(PNPhPR)] (R=Cy (2g), Ph (2a), Et (2c)). Figure 4. 31P NMR spectrum of the fac-[Mo(CO)3(PNPhPEt)] complex (2 c) including traces of the meridional isomer at 48.9 ppm. Wiley VCH Montag, 28.08.2023 2399 / 317727 [S. 4/15] 1 Eur. J. Inorg. Chem. 2023, e202300280 (4 of 14) © 2023 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202300280 10990682c, 0, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202300280 by Cochrane Germany, Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 54 4.1 Molybdenum tricarbonyl complexes supported by linear PNP ligands: Influence of Pand N-substituents on structure, stability and on the activation of small molecules and outer lines with respect to lines 2 and 5 correspond to ɛ and 1 - ɛ, respectively (cf. Figure 5, bottom), with the latter given by Equation (2):[28,33] D� ¼ 1 2ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi nAnB ð Þ �0:5 JAX JBX ð Þ½ �2þJ2 AB q(1) 1e¼ 1 4JAX JBX ð Þ2 1 4JAXJBX ð Þ2þJ2 AB (2) Based on these equations, approximate values for the coupling constants JAX, JBX and JAB as well as the chemical shift difference νA–νBwere determined and subsequently refined by fitting to the experimental spectra (see Figures S62-S65). The shifts and coupling parameters found for the carbonyl and phosphine groups of complexes 2a,2 c,2e, and 2f are collected in Table 3. In addition to similar chemical shifts of the CO ligands, the coupling constants to the P-nuclei are also very similar, indicating not only similar geometries but also comparable electronic structures. Unfortunately, complex 2g showed such a weak solubility in most NMR solvents that it was impossible to obtain spectra of sufficient quality to derive these parameters. Besides the ABX patterns of the equatorial CO ligands, the 13C NMR signals of the ethylene bridges (Figure 5c and 5d) and terminal ethyl groups (Figure 5e and 5f) also reflect coupling to the phosphorus atoms. With decreasing spin-spin coupling of the 13C and 31P nuclei, the difference between D+and Ddecreases, leading to a smaller distance between the inner lines (3 & 4). Moreover, the intensity of the outer lines (1 & 6) decreases, which in turn increases the intensity of the inner lines. This is, e.g., visible for the signal of the ethyl bridge proximal to the phosphorus atom (Figure 5d). Ultimately, the six-line pattern collapses into a triplet (Figure 5c; arrows indicate the positions of the very weak outer lines). Similar spectra are also observed for the other complexes (Figure S35, S43, S47, S51, and S53). The derived 13C NMR parameters are collected in Table 3. Thermodynamics of Metal to PNP-Ligand Bonding Data obtained from NMR, IR, and Raman spectroscopy indicate a facial geometry of complexes 2a,2c,2e,2f, and 2 g in solution and in the solid state. The meridional isomers are only visible as minor traces (<1%) (Figure 4) in the 31P NMR spectra. In order to understand the preference of the ligands for facial coordination, a DFT study was conducted. To this end ligand exchange reactions with the different ligands were treated theoretically, considering both mer and fac configurations for the entire series of complexes. This allows to directly compare their relative stabilities (Figure 6, Table S8 & S9), leading to the following conclusions: (i) in agreement with the results of Keskin et al., the facial isomer of [Mo(CO)3(PNPhPPh] (2a) is more favorable than its mer counterpart; (ii) a thermodynamically more stable tricarbonyl complex (2c) is generated by the PNPhPEt ligand whereby the fac isomer is also at lower energy than its mer analogue; (iii) for all PNPhPRligands with terminal alkyl substituents exhibiting a higher steric demand than ethyl, the stability of the derived Mo(CO)3complex decreases and mer is at lower energy than fac (Figure 6). Notably, both isomers of 2h with di-tert-butylphosphine groups are very unfavorable. In the facial isomer, the bond lengths are too elongated to consider it a possible product; accordingly, the calculated energy is very high. The meridional geometry of 2h is more favorable, but still at high energy compared to all other complexes (cf. θO(PtBu3)=167.1°).[34] This Figure 5. Top: 13C NMR spectrum of [Mo(CO)3(PNPhPEt)] (2c) with the multiple ABX patterns of the different 13C atoms (a-f) found due to the coupling to the phosphorus nuclei. The signals of NPh group are marked with an asterisk (*) and signals of solvents with two asterisks (**). Arrows indicate outer lines of little intensity. Bottom: Six-line pattern emerging from ABX-spectrum. Wiley VCH Montag, 28.08.2023 2399 / 317727 [S. 5/15] 1 Eur. J. Inorg. Chem. 2023, e202300280 (5 of 14) © 2023 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202300280 10990682c, 0, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202300280 by Cochrane Germany, Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 55 Chapter 4 Project 2 may explain the experimental finding this complex could not be synthesized (see above). Although the calculations overall suggest a preference for the meridional isomers for all Mo(CO)3complexes supported by PNPhPRligands with sterically more demanding substituents than ethyl (cf. Figure 6), only the facial isomers were obtained experimentally (see above). In order to understand this apparent contradiction to the theoretically calculated energetics, we compared our findings with results obtained on similar compounds in the literature. Notably, Keskin et al. also reported that only facial [Mo(CO)3(PNPhPPh)] (2a) was obtained initially; the meridional isomer only formed in a small amount when the facial complex was stored in solution for weeks.[26] However, [Mo(CO)3(PNHPR)] complexes with PNHPRligands (R= iPr, Cy, tBu) containing a central NH donor instead of NPh seem to behave differently.[35,36] For [Mo(CO)3(PNHPiPr)]; e.g., a crystal structure with facial geometry was obtained,[35] but the corresponding NMR and IR spectra reflect a meridional geometry (Supp. Mat. of ref. [35]). Furthermore, [Mo- (CO)3(PNHPCy)] is found to exist in, both, meridional and facial geometry in solution, with the ratio of the two isomers being solvent-dependent.[36] In order to identify possible differences in the stabilities of PNPhP and PNHP complexes, analogous isodesmic calculations of ligand exchange reactions as performed for the former (Figure 6) were conducted for the latter systems as well (Figure S59 & S60). In agreement with the PNPhP complexes, the fac isomers tend to become more unfavorable with increasing steric demand of the phosphine residues, accounting for the experimentally observed formation of mer isomers (along with their fac counterparts). The lack of observation of mer isomers in case of the PNPhP complexes thus has to to be of kinetic, not thermodynamic origin. Theoretical information on possible mer-fac isomerisations of [Mo(CO)3(PNP)] complexes in solution can be obtained by DFT calculations as well. In principle, these processes can occur in a non-dissociative fashion via distorted octahedral transition states (Figure 7, Figure S65 & S66). Interestingly, calculations of the corresponding energy barriers for model complexes with PMe2groups (PNHPMe), reveal that the transition states for the amine PNHPMe and phenylamine PNPhPMe systems (Figure 7, TSISONRCO3) are unfavorable. The corresponding energy barriers of around 30 kcal/mol make such transformations unlikely. Additionally, no clear differences between the NH and NPh ligands were found, which contrasts with the experimental observations, showing significantly different behaviour. Another process possibly enabling an isomeric transformation involves the preliminary dissociation of a CO ligand from the 18-electron [Mo(CO)3(PNRP)] complexes, leading to pentacoordinate square-pyramidal 16-electron complexes [Mo- (CO)2(PNRP)] (Figure 8 and 9, Figure S67 & S68, Table S10).[26,37] The resulting empty coordination site makes it energetically much more favorable for the ligand to undergo a conformational rearrangement. Notably, this dissociative substitution mechanism has been experimentally evidenced for [Mo- (CO)3(PNPhPPh)] (2a) by Keskin et al.[26] Table 3. Chemical shift and coupling constants of the equatorial carbonyl ligands of the fac-[Mo(CO)3(PNPhPR)] complexes. Complex δ(13COax) [ppm] δ(13COeq) [ppm] δ(31P) [ppm] j2JAB j [Hz] 2JAX (trans) [Hz] 2JBX (cis) [Hz] ~νAB [Hz] [Mo(CO)3(PNPhPPh)] (2a) 231.6 219.6 38.0 21.6 39.4 14.0 1.30 [Mo(CO)3(PNPhPEt)] (2c) 231.4 220.3 33.7 29.1 32.4 10.8 1.80 [Mo(CO)3(PNPhPCyp)] (2e) 232.0 220.8 41.6 22.7 36.5 13.7 0.97 [Mo(CO)3(PNPhPiPr)] (2f) 231.9 220.8 51.6 26.5 33.7 8.8 1.69 [Mo(CO)3(PNPhPCy)] (2g) 232.1 221.0 40.8 – – – – Figure 6. Theoretical relative Gibbs energies (including solvation) of the facial and meridional [Mo(CO)3(PNPhPR)] complexes with different phosphine groups (R=Et (2c), Ph (2 a), Cyp (2 e), iPr (2f), Cy (2g), tBu (2h)), relative to the most stable complex mer-2c and calculated by ligand exchange reactions (PBE0/def2-TZVPP). Wiley VCH Montag, 28.08.2023 2399 / 317727 [S. 6/15] 1 Eur. J. Inorg. Chem. 2023, e202300280 (6 of 14) © 2023 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202300280 10990682c, 0, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202300280 by Cochrane Germany, Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 56 4.1 Molybdenum tricarbonyl complexes supported by linear PNP ligands: Influence of Pand N-substituents on structure, stability and on the activation of small molecules The energies of the corresponding dissociative intermediates [Mo(CO)2(PNRPMe)] (R=H, Ph) indicate that only the CO ligands trans to P or CO coligands dissociate, leading to squarepyramidal isomers with the amine donor in equatorial position (Figure 8 and 9, fac1NH, mer1NH, mer2NH, fac1NPh and mer2NPh). Therefore, only these are taken into account for the subsequent rebinding of CO and indeed, a significantly different behavior is observed for the NH and NPh ligands presented here: In the case of [Mo(CO)2(PNHPMe)], two equally stable mer isomers are found (Figure 8, mer1NH and mer2NH) which are in equilibrium with each other via a very low-lying trigonal bipyramidal transition state (Figure 8, TSMERNHCO2). This makes both isomers equally available for conversion into a facial isomer. The actual isomeric transformation then occurs between mer1NH and fac1NH via a distorted trigonal bipyramidal transition state (TSISONHCO2). Due to the relatively low energy barrier of 11 kcal/mol from mer1NH or 3 kcal/mol from fac1NH,mer and fac complexes are in an equilibrium in solution, which is in accordance with the experimental observations in the literature.[36] In case of [Mo(CO)2(PNPhPMe)], only one meridional and one facial isomer is stable due to an agostic hydrogen bond between the phenyl group and the molybdenum center, which occupies the vacant coordination site (mer1NPh and fac1NPh, Figure 9). This and the steric demand of the migrating NPh group lead to a much higher energy barrier of 18 kcal/mol from mer1NPh or 9 kcal/mol from fac1NPh complexes, making this process either impossible or at least very slow. Therefore, the initial geometry of the complex resulting from reaction of the precursor with the PNP ligand is conserved. For fac-[MoFigure 7. Theoretical relative Gibbs energies in solution of the isomeric transformation of facand mer-[Mo(CO)3(PNRPMe)] model complexes (R=H (red), Ph (blue)) via transition state TSISONRCO3 (PBE0/def2-TZVPP(D3BJ)) (P=PMe2) Figure 8. Theoretical relative Gibbs energies in solution of the isomers of pentacoordinated model complex [Mo(CO)2(PNHPMe)], including the isomeric transformation via transition state TSISOHand the conversion of the two possible meridional isomers via TSmer (PBE0/def2-TZVPP(D3BJ)) (P=PMe2). Wiley VCH Montag, 28.08.2023 2399 / 317727 [S. 7/15] 1 Eur. J. Inorg. Chem. 2023, e202300280 (7 of 14) © 2023 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202300280 10990682c, 0, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202300280 by Cochrane Germany, Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 57 Chapter 4 Project 2 (CO)3(cht)], which has been used in this study for complexation of the PNPhPRligands, it has been established that reaction with linear tridentate ligands exclusively leads to fac-complexes.[38,39] Due to the high barriers, conversion to possibly lower-lying mer-isomers thus is hindered, explaining the lack of observation of mer-isomers. Conclusions In order to study the influence of the different phosphine groups on the coordination behavior of PNP ligands and examine their impact on the activation small molecules like CO, a series of tridentate PNPhPRligands (R=Ph, Me, Et, Pln, Cyp, iPr, Cy, tBu) and corresponding molybdenum tricarbonyl complexes [Mo(CO)3(PNPhPR)] (R=Ph, Et, Cyp, iPr, Cy,) were prepared and characterized by NMR and vibrational spectroscopy. In case of [Mo(CO)3(PNPhPEt)] the determination of the crystal structure was possible as well. Upon replacing the PNPhPPh ligand of the parent complex [Mo(CO)3(PNPhPPh)] by ligands with alkylphosphine groups the activation of the bound carbonyl ligands was found to increase. Within the series of new complexes supported by PNPhPR ligands with alkyl substituents, however, differences in CO stretching frequencies and, thus, CO activation, were small. Nevertheless, the relative stabilities of these complexes significantly decrease with increasing steric bulk of the terminal alkyl substituents R. This becomes evident from isodesmic calculations, replacing, e.g., the PNPhPPh ligand of the parent [Mo- (CO)3(PNPhP)] complex by corresponding ligands with PR2 groups (R=Et, Cyp, iPr, Cy). This way, it becomes clear that the thermodynamically most stable member of the entire family of PNPhPRcomplexes is [Mo(CO)3(PNPhPEt)], exhibiting strongly electron-donating PEt2groups with little steric bulk. On the other hand, these calculations provide an explanation for the fact that the complex with the sterically most demanding tertbutyl substituents could not be obtained. Experimental and theoretical information was also obtained on the preferred stereochemical configuration (mer vs. fac) of the [Mo(CO)3(PNPhPR)] complexes. Surprisingly, all of the synthesized [Mo(CO)3(PNPhPR)] complexes exhibit the fac geometry, irrespective of the electronic and steric properties of the terminal phosphine groups. This is, e.g., evident from NMR spectroscopy where the meridional isomers were only found as minor byproducts visible in small amounts (<1%) in the 31P spectra. Moreover, the preferred formation of the fac isomers could also be inferred from X-ray structure determination and vibrational spectroscopy where typical CO splitting patterns were observed. Additional splittings present in the IR and Raman spectra were attributed to factor group or Davydov splittings. For the parent [Mo(CO)3(PNPhPPh)] complex (2a) the preference for the fac-isomer had already been evidenced by Keskin et al. Notably, they were able to generate the fac isomer from the complex [Mo(CO)4(k2-PNPhP)] with a yield of <1%. Furthermore, they found that the mer isomer is formed in small amounts upon letting a solution of the fac isomer stand for a long time. From their observations, they concluded that the mer-isomer is formed from its fac counterpart by a fac!mer isomerization process proceeding via a dissociative pathway. This scenario has also been advanced by Crabtree regarding the stereoisomerism of [Mo(CO)3L3] complexes.[37] Theoretical support for the hypothesis that the fac-isomer of the complex [Mo(CO)3(PNPhPPh)] is more stable than its mer counterpart was obtained by Keskin et al. from DFT calculations.[26] Notably, Siclovan et al. synthesized the [M- (CO)3(PNMeP)] (M=Mo, W) complex and also found the facial isomer to be more stable than its meridional analog.[7] In this context it has to be noted that a general electronic-structural preference for the fac geometry in Mo(0) tricarbonyl complexes results from the fact that all three t2g are able to interact in an equal fashion with the CO ligands in a π-backbonding manner whereas this is not possible for the mer configuration. However, this electronic effect should potentially be counteracted by a steric effect if the residues on the terminal phosphine donors become more bulky than phenyl groups. For sterically very demanding substituents, this ultimately may lead to a preference of the mer configuration. This is in fact observed for molybdenum tricarbonyl complexes supported by PNHP ligands,[35,36] but apparently does not apply to our systems. In order to address this problem, calculations of the relative stabilities of mer-fac isomers have been performed for the Figure 9. Theoretical relative Gibbs energies in solution of the different isomers of pentacoordinated model complex [Mo(CO)2(PNPhPMe)], including isomeric transformation via transition state (TSISOPh) (PBE0/def2-TZVPP(D3BJ)) (P=PMe2). Wiley VCH Montag, 28.08.2023 2399 / 317727 [S. 8/15] 1 Eur. J. Inorg. Chem. 2023, e202300280 (8 of 14) © 2023 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202300280 10990682c, 0, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202300280 by Cochrane Germany, Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 58 4.1 Molybdenum tricarbonyl complexes supported by linear PNP ligands: Influence of Pand N-substituents on structure, stability and on the activation of small molecules entire series of complexes. While the thermal stability of these complexes overall decreases with increasing steric bulk of the substituents (see above), these calculations also reveal that a preference for the fac isomer is restricted to the PNPhPPh complex and the PNPhPEt complex which exhibits the least sterically demanding alkyl substituents. For all complexes with bulkier residues the mer isomers are thermodynamically favored. However, this is not observed experimentally. We thus suspected that formation of the mer products might be kinetically hindered in all of these complexes. In order to check this hypothesis, potential fac-mer isomerization processes have been treated with the help of DFT. Notably, calculations of the model systems [Mo(CO)3(PNRPMe)] (R=H, Ph) showed that a direct isomeric transformation of these hexacoordinated complexes is unlikely. Nevertheless, it can occur after preliminary dissociation of CO, in the course of a dissociative process (see above). Our calculations indicate that such a process is thermally allowed in the PNHP systems, but is hindered in the PNPhP systems due to the steric demand of the phenyl group and an agostic interaction present both in the facial and the meridional isomers involved in the transformation. This acts to significantly increase the energetic barrier for isomerization in PNPhP systems, making it very slow or even suppressing it completely. The stereoisomeric configuration of the primary products resulting from reaction of the Mo(CO)3 precursor with the PNPhP ligands thus is more or less retained. As it has been established that reaction of the precursor fac- [Mo(CO)3(cht)] with linear tridentate ligands exclusively leads to fac products,[38,39] it becomes understandable that all of the investigated complexes 2a,2 c,2e,2f, and 2g exhibit the fac geometry and no mer isomers have been observed. This stereochemical peculiarity of the PNPhP ligands may entail characteristic differences in the reactivity of derived transitionmetal complexes to analogues supported by PNHP ligands, comparable to those evidenced before in the context of CO2 hydrogenation catalyzed by Ru(PNHP) and -PNPhP complexes.[19] Experimental Section Materials and methods: All syntheses containing airand moisturesensitive compounds were carried out under a nitrogen atmosphere using Schlenk techniques. All solvents were dried under argon atmosphere prior to use. Commercially available substances were used as purchased without further purification. The chemicals used for synthesis were obtained from Sigma-Aldrich Co., abcr GmbH & Co. KG., and Fisher Scientific GmbH. Spectroscopy: The NMR spectra were measured on a Bruker Avance III HD 400 pulse Fourier Transform spectrometer at 400.13 MHz (1H), 100.61 MHz (13C), and 161.98 MHz (31P). The spectra were either referenced to solvent residue (5.32 ppm for CD2Cl2) or TMS (δ1H= 0 ppm), 85% H3PO4(δ31P=0 ppm) as substitutive standards. The IR spectra were measured on a Bruker alpha FT-IR spectrometer and the Raman spectra on Bruker Vertex70 FT-IR with RAMII module. The elemental analyses were performed using a VarioMICRO cube element analyzer. Single Crystal Structure Determination: Data collection was performed with an XtaLAB Synergy, Dualflex diffractometer with a microfocus tube using CuKαradiation (λ=1.54184). The structure was solved with SHELXT[40] using Intrinsic Phasing and refined with SHELXL[41] using Least Squares minimisation. All non-hydrogen atoms were refined anisotropically. The CH H atoms were positioned with idealized geometry (methyl H atoms allowed to rotate but not to tip) and were refined isotropic with Uiso(H)= 1.2 Ueq(C) (1.5 for methyl H atoms using a riding model. Deposition Number 2156795 contains the supplementary crystallographic data for this paper. These data are provided free of charge by the joint Cambridge Crystallographic Data Centre and Fachinformationszentrum Karlsruhe Access Structures service. Computational details: All calculations were performed using the ORCA 4.2.1 program package[42] with PBE0 functional[43] and def2TZVPP basis set.[44] Furthermore, Grimmes dispersion correction[45] with Becke-Johnson damping (D3BJ),[46] density fitting approximation (RIJCOSX)[47] and solvation correction (CPCM) were applied.[48] All minimum structures were optimized to have no imaginary frequencies. The transition states were optimized with exactly one imaginary frequency. General procedures and synthesis: The syntheses of dimethyl phosphine,[49] diisopropyl phosphine,[50] phospolane,[51] NN-bis(2chloroethyl)aniline[52] and N,N-bis(2-(ptoluenelsulfonyl)ethyl)aniline[25] were prepared according to literature. General procedure A: Synthesis of the PNP ligands. 1 eq. of NNbis(2-chloroethyl)aniline was dissolved in thf and cooled to 0°C. In a second vessel, the phosphine (2.1 eq.) was also dissolved in THF, cooled to 0°C and n-butyllithium solution (2.5 mol/L) was added dropwise over 20 min. The solution was stirred for 30 min at 0°C. Within 15 min the NN-Bis(2,2-dichloroethyl)aniline solution was added to the phosphine/n-butyllithium solution at 0°C. The solution was stirred for 18 h at room temperature. 0.5 mL degassed water was added and the solvent was evaporated in vacuo. The residue was resolved in n-pentane (30 mL) and filtered through Celite® and basic aluminum oxide. General procedure B: Synthesis of the molybdenum tricarbonyl complexes. 1 eq. of [Mo(CO)3(cht)] was dissolved in 3 mL toluene giving a red solution. The PNP-Ligand was dissolved in 2 mL toluene and added to the precursor complex solution. The solution was stirred for 18 h in the glovebox. The resulting crude product was filtered and washed with n-pentane (5 mL). The product was dried in vacuo resulting in a yellowish solid. NN-Bis(2-(diphenylphosphino)ethyl)aniline (PNPhPPh, 1 a) Following general procedure A, the PNPhPPh (1a) ligand was obtained as colourless solid (796 mg, 1.54 mmol, 67%). 31P{1H} NMR: (161.98 MHz, CD2Cl2, 300 K): δ=-20.1 (s, 2 P, PPh2) ppm. 1H NMR (400.13 MHz, CD2Cl2, 300 K): δ=7.35–7.30 (m, 8 H, P- (CH)phenyl), 7.26–7.22 (m, 12 H, P-(CH)phenyl), 7.03–6.99 (m, 2 H, N- (CH)phenyl), 6.56–6.52 (m, 1 H, N-(CH)phenyl), 6.28–6.26 (m, 2 H, N- (CH)phenyl), 3.29–3.24 (m, 4 H, N-CH2), 2.22–2.18 (m, 4 H, P-CH2) ppm. 13C{1H} NMR (100.62 MHz, CD2Cl2, 300 K): δ=145.7 (s, 1 C, N- (Ci,phenyl)), 139.9 (d, JPC =12.2 Hz, 4 C, P-(Ci,phenyl)), 131.7 (d, JPC = 18.8 Hz 8 C, P-(Co,phenyl)), 128.2 (s, 2 C, N-(Co,phenyl)), 127.7 (s, 4 C, P- (Cp,phenyl)), 127.4 (d, JPC =6.80 Hz 8 C, P-(Cm,phenyl)) 115.4 (s, 1 C, Cp,phenyl), 111.2 (s, 2 C, Cm,phenyl), 46.6 (d, 2JPC =25.6 Hz, 2 C, N-CH2), 25.0 (d, JPC =14.4 Hz, 2 C, P-CH2) ppm. IR (ATR): ~ v=3068 (w), 3049 (w), 3025 (w), 3018 (sh), 3001 (w), 2961 (w), 2928 (w), 2902 (w). 2869 (sh), 2855 (w), 1969 (vw), 1957 (vw),1914 (vw), 1903 (vw), 1886 (vw), 1808 (w), 1764 (vw), 1595 (s), Wiley VCH Montag, 28.08.2023 2399 / 317727 [S. 9/15] 1 Eur. J. Inorg. Chem. 2023, e202300280 (9 of 14) © 2023 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202300280 10990682c, 0, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202300280 by Cochrane Germany, Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 59 Chapter 4 Project 2 1571 (m), 1501 (s), 1476 (s), 1461 (w), 1429 (s), 1414 (w), 1389 (m), 1359 (w), 1352 (m), 1337 (w), 1319 (vw), 1305 (w), 1275 (sh), 1260 (s), 1218 (w), 1195 (s), 1180 (w), 1148 (m), 1094 (m), 1064 (w), 1039 (w), 1026 (w), 1011 (w), 995 (w), 984 (w), 947 (w), 913 (w), 907 (vw), 865 (m), 815 (sh), 794 (s), 734 (vs), 688 (vs), 667 (sh), 614 (vw), 554 (m), 528 (m), 504 (vs), 476 (vs), 430 (m), 411 (sh) cm1. NN-Bis(2-(dimethylphosphino)ethyl)aniline (PNPhPMe, 1 b) Following general procedure A, the PNPhPMe (1b) ligand was synthesized, however, the phosphine solution was cooled to 78°C instead of 0°C as described in the procedure. The ligand (1 b) was obtained as colourless oil (236.7 mg, 0.88 mmol, 63%). 31P{1H} NMR: (161.98 MHz, CD2Cl2, 300 K): δ=-54.9 (s, 2 P, PMe2) ppm. 1H NMR (400.13 MHz, CD2Cl2, 300 K): δ=7.22–7.17 (m, 2 H, CHphenyl), 6.68–6.61 (m, 3 H, CHphenyl), 3.45–3.39 (m, 4 H, CH2), 1.70–1.65 (m, 4 H, CH2), 1.06 (d, 2JPH =2.29 Hz, 12 H, CHmethyl) ppm. 13C{1H} NMR (100.62 MHz, CD2Cl2, 300 K): δ=147.4 (s, 1 C, Ci,phenyl), 129.2 (s, 2 C, Co,phenyl), 115.7 (s, 1 C, Cp,phenyl), 112.2 (s, 2 C, Cm,phenyl), 47.7 (d, 2JPC =21.4 Hz, 2 C, N-CH2), 29.8 (d, 1JPC =12.6 Hz, 2 C, P-CH2), 13.8 (d, JPC =13.1 Hz, 4 C, P-CH3) ppm. IR (ATR): ~ v=3090 (w), 3058 (w), 3038 (w), 3023 (w), 2951 (s), 2923 (m), 2892 (s), 2853 (sh), 2812 (w), 1598 (s), 1574 (sh), 1504 (s), 1461 (vw), 1449 (vw), 1428 (sh), 1420 (m), 1394 (w), 1354 (s), 1292 (sh), 1278 (m), 1213 (w), 1186 (m), 1159 (vw), 1142 (sh), 1128 (m), 1084 (vw), 1042 (m), 1007 (br m), 990 (w), 967 (vw), 938 (s), 885 (m), 818 (w), 747 (s), 710 (w), 694 (s), 668 (w), 531 (vw), 508 (w), 434 (vw), 416 (vw) cm-1. NN-Bis(2-(diethylphosphino)ethyl)aniline (PNPhPEt, 1 c) Following general procedure A, the PNPhPEt (1c) ligand was obtained as colourless oil (411 mg, 1.26 mmol, 25%). 31P{1H} NMR: (161.98 MHz, CD2Cl2, 300 K): δ=-25.7 (s, 2 P, PEt2) ppm. 1H NMR (400.13 MHz, CD2Cl2, 300 K): δ=7.20–7.16 (m, 2 H, CHphenyl), 6.67–6.60 (m, 3 H, CHphenyl), 3.45–3.40 (m, 4 H, N-CH2), 1.70–1.65 (m, 4 H, P-CH2), 1.48–1.42 (m, 8 H, P-CH2-CH3), 1.11–1.03 (m, 12 H, PCH2-CH3) ppm. 13C{1H} NMR (100.62 MHz, CD2Cl2, 300 K): δ=147.4 (s, 1 C, Ci,phenyl), 129.3 (s, 2 C, Co,phenyl), 115.8 (s, 1 C, Cp,phenyl), 112.3 (s, 2 C, Cm,phenyl), 48.3 (d, 2JPC =22.6 Hz, 2 C, N-CH2), 24.3 (d, 1JPC =15.9 Hz, 2 C, P-CH2) 19.0 (d, 1JPC =18.9 Hz, 4 C, P-CH2-Me), 13.8 (d, 2JPC =12.6 Hz, 4 C, -CH3) ppm. IR (ATR): ~ v=3085 (w), 3043 (w, br), 2955 (w), 2923 (s), 2877 (s), 2821 (sh), 2821 (sh), 2220 (w), 2135 (w), 1600 (s), 1505 (s), 1436 (s), 1400 (m), 1368 (s), 1210 (w), 1200 (vw) 1158 (s), 1131 (s), 989 (m), 967 (vw), 842 (sh), 789 (sh), 747 (vs), 695 (s), 619 (vw), 567 (w), 513 (w), 451 (vw) cm-1. NN-Bis(2-phospholanoethyl)aniline (PNPhPln, 1 d) NN-Bis(2-chloroethyl)aniline (320 mg, 147 mmol) was dissolved in THF (10 mL). In a second vessel lithium phospholanide (500 mg, 2.95 mmol) was dissolved in 8 mL THF and cooled to 0°C. 1.2 mL (3.00 mmol) of n-butyllithium solution was added dropwise and stirred in the cold for 40 min. The aniline solution was added slowly to the phosphine/n-buli solution and it was stirred for 4 d at room temperature. All volatile components were removed in vacuo and the residue was extracted with n-pentane and dichloromethane before filtering through celite and basic allox. The solvents were removed yielding a yellow oil as product (233 mg, 0.72 mmol, 49%). 31P{1H} NMR: (161.98 MHz, CDCl3, 300 K): δ=-32.0 (s, 2 P, Pln) ppm. 1H NMR (400.13 MHz, CDCl3, 300 K): δ=7.24–7.20 (m, 2 H, CHphenyl), 6.68–6.64 (m, 3 H, CHphenyl), 3.41–3.36 (m, 4 H, N-CH2), 1.85–1.65 (m, 4 H, P-CH2, 8 H, (CH2)Pln), 1.63–1.59 (m, 4 H, (CH2)Pln), 1.53–1.46 (m, 4 H, (CH2)Pln), ppm. 13C{1H} NMR (100.62 MHz, CDCl3, 300 K): δ=146.1 (s, 1 C, Ci,phenyl), 128.3 (s, 2 C, Co,phenyl), 114.9 (s, 1 C, Cp,phenyl), 111.2 (s, 2 C, Cm,phenyl), 47.7 (d, 2JPC =21.4 Hz, 2 C, N-CH2), 26.8 (d, 1JPC =3.8 Hz, 4 C, (CH2)Pln), 25.8 (d, JPC =18.6 Hz, 4 C, P-CH2), 25.0 (d, 1JPC =11.0 Hz, 4 C, (CH2)Pln) ppm. IR (ATR): ~ v=3090 (w), 3057 (w), 3038 (w), 3023 (w), 2929 (s), 2873 (vw), 2855 (s), 2823 (sh), 2675 (vw), 1916 (w), 1905 (w), 1600 (vs), 1570 (sh), 1502 (vs), 1457 (w), 1445 (m), 1352 (s), 1322 (sh), 1298 (vw), 1275 (sh), 1256 (s), 1209 (vw), 1183 (m), 1157 (vw), 1118 (sh), 1107 (s), 1087 (sh), 1054 (vw), 1039 (w), 1013 (s), 950 (vw), To2 (vw), 859 (w), 828 (sh), 792 (s), 743 (vs), 688 (vs), 657 (vw), 620 (vw), 561 (vw), 508 (m), 466 (vw), 398 (m) cm1. NN-Bis(2-(dicyclopentylphosphino)ethyl)aniline (PNPhPCyp, 1 e) Following general procedure A, the PNPhPCyp (1e) ligand was obtained with the exception that the solution was stirred for 4 d yielding the product as colourless solid (540 mg, 1.11 mmol, 76%). 31P{1H} NMR: (161.98 MHz, CDCl3, 300 K): δ=-7.93 (s, 2 P, PCp2) ppm. 1H NMR (400.13 MHz, CDCl3, 300 K): δ=7.24–7.20 (m, 2 H, CHphenyl), 6.68–6.63 (m, 3 H, CHphenyl), 3.49–3.44 (m, 4 H, N-CH2), 1.97–1.85 (m, 8 H, CH2, Cp), 1.74 (m, 2JPH =2.35 Hz, 3JHH =8.79 Hz, 4 H, P-CH2), 1.70– 1.63 (m, 8 H, CH2, Cp), 1.62–1.51 (m, 8 H, CH2, Cp), 1.48–1.33 (m, 8 H, CH2, Cp) ppm. 13C{1H} NMR (100.62 MHz, CDCl3, 300 K): δ=147.2 (s, 1 C, Ci,phenyl), 129.5 (s, 2 C, Co,phenyl), 115.8 (s, 1 C, Cp,phenyl), 112.2 (s, 2 C, Cm,phenyl), 49.2 (d, 2JPC =40.6 Hz, 2 C, N-CH2), 36.2 (d, 2JPC =10.3 Hz, 4 C, P-CCp) 31.3 (d, 1JPC =16.9 Hz, 4 C, P-CCp), 30.5 (d, 2JPC =12.5 Hz, 4 C, P-CCp), 26.6 (d, 3JPC =7.40 Hz, 4 C, P-CCp), 26.3 (d, 3JPC =6.30 Hz, 4 C, P-CCp), 22.8 (d, 1JPC =17.9 Hz, 2 C, P-CCp) ppm. IR (ATR): ~ v=3211(w, br), 3089 (w), 3060 (w), 3036 (w), 3018 (w), 2942 (vs), 2902 (w), 2858 (vs), 2056 (vw), 1904 (vw), 1800 (vw), 1752 (vw), 1602 (s), 1595 (s), 1566 (m), 1506 (vs), 1466 (sh), 1455 (sh), 1444 (s), 1402 (s), 1350 (s), 1320 (vw), 1283 (s), 1260 (w), 1215 (w), 1184 (vs), 1155 (vw), 1126 (s), 1086 (sh), 1059 (vw), 1034 (m), 1002 (s), 980 (w), 942 (vw), 921 (vw), 904 (m), 853 (m), 800 (m), 755 (sh), 738 (vs), 703 (sh), 689 (vs), 545 (m), 508 (s), 490 (vw), 457 (vw), 423 (w) cm1. NN-Bis(2-(diisopropylphosphino)ethyl)aniline (PNPhPiPr, 1 f) Diisopropylphosphine (0,67 g, 5.67 mmol) was solved in 10 mL THF and cooled to 0°C. n-Butyllithium solution in hexane (2.4 mL, 6.00 mmol) was added dropwise and stirred for 30 min at 0°C. Afterwards, the solution was stirred for an additional 30 min at room temperature. The solution turned orange. NN-Bis(2-(ptoluenelylsulfonyl)ethyl)aniline (1.26 g, 2.57 mmol) was dissolved in 10 mL THF and both solutions were cooled to 0°C. The phospine/nbutyllithium solution was added dropwise to the aniline giving a yellow solution. After an hour, the solution turned green and red/ orange after 18 h. The reaction was quenched with degassed water (5 mL) and all volatile components were removed in vacuo. The Wiley VCH Montag, 28.08.2023 2399 / 317727 [S. 10/15] 1 Eur. J. Inorg. Chem. 2023, e202300280 (10 of 14) © 2023 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202300280 10990682c, 0, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202300280 by Cochrane Germany, Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 60 4.1 Molybdenum tricarbonyl complexes supported by linear PNP ligands: Influence of Pand N-substituents on structure, stability and on the activation of small molecules residue was resolved in 40 mL diethyl ether and filtered through Celite® and basic aluminium oxide. The PNPhPiPr (1f) ligand was obtained as colourless oil (411 mg, 1.26 mmol, 25%). 31P{1H} NMR: (161.98 MHz, CD2Cl2, 300 K): δ=1.53 (s, 2 P, PiPr2) ppm. 1H NMR (400.13 MHz, CD2Cl2, 300 K): δ=7.24–7.20 (m, 2 H, CHphenyl), 6.68–6.62 (m, 3 H, CHphenyl), 3.51–3.46 (m, 4 H, N-CH2), 1.79 (dsept., 2JPH =2.11 Hz, 3JHH =7.11 Hz, 4 H, P-CH), 1.71–1.65 (m, 4 H, P-CH2), 1.15–1.08 (m, 24 H, P-CH-CH3) ppm. 13C{1H} NMR (100.62 MHz, CD2Cl2, 300 K): δ=147.2 (s, 1 C, Ci,phenyl), 129.3 (s, 2 C, Co,phenyl), 115.6 (s, 1 C, Cp,phenyl), 112.1 (s, 2 C, Cm,phenyl), 50.4 (d, 2JPC =32.2 Hz, 2 C, N-CH2), 23.4 (d, JPC =12.3 Hz, 4 C, P-CH) 20.0 (d, 2JPC =16.3 Hz, 4 C, P-CH-Me), 19.8 (d, JPC =12.6 Hz, 2 C, CH2-P), 18.7 (d, 2JPC =9.40 Hz, 4 C, P-CH-Me) ppm. IR (ATR): ~ v=3091 (w), 3061 (w), 3038 (w), 2958 (sh), 2945 (vs), 2923 (w), 2887 (w), 2861 (vs), 1914 (vw), 1595 (vs), 1571 (vw), 1500 (vs), 1456 (s), 1382 (m), 1361 (sh), 1350 (s), 1277 (m), 1235 (vw), 1214 (w), 1183 (m), 1158 (w), 1127 (m), 1099 (vw), 1086 (sh), 1035 (s), 1014 (w), 993 (m), 919 (m), 887 (m), 859 (w), 806 (w), 742 (vs), 716 (w), 693 (s), 649 (w), 619 (w), 572 (vw), 509 (w), 466 (w) cm1. NN-Bis(2-(dicyclohexylphosphino)ethyl)aniline (PNPhPCy, 1 g) Following general procedure A, the PNPhPCy (1 g) ligand was obtained as colourless solid (2.00 mg, 3.69 mmol, 99%). 31P{1H} NMR: (161.98 MHz, CD2Cl2, 300 K): δ=-8.21 (s, 2 P, PCy2) ppm. 1H NMR (400.13 MHz, CD2Cl2, 300 K): δ=7.21–7.17 (m, 2 H, CHphenyl), 6.63–6.59 (m, 3 H, CHphenyl), 3.44–3.38 (m, 4 H, N-CH2), 1.79–1.63 (m, 24 H, CH2), 1.60–1.53 (m, 4 H, P-CH2), 1.32–1.14 (m, 20 H, CH2) ppm. 13C{1H} NMR (100.62 MHz, CD2Cl2, 300 K): δ=147.1 (s, 1 C, Ci,phenyl), 129.2 (s, 2 C, Co,phenyl), 115.4 (s, 1 C, Cp,phenyl), 112.1 (s, 2 C, Cm,phenyl), 50.4 (d, 2JPC =33.2 Hz, 2 C, N-CH2), 33.3 (d, JPC =12.7 Hz, 4 C, CH2), 30.4 (d, 2JPC =14.9 Hz, 4 C, CH2), 29.1 (d, JPC =7.93 Hz, 4 C, CH2), 27.3 (d, JPC =19.5 Hz, 4 C, CH2), 27.3 (s, 4 C, CH2), 26.6 (s, 4 C, CH2), 19.5 (d, JPC =20.3 Hz, 2 C, P-CH2) ppm. IR (ATR): ~ v=3087 (vw), 3058 (w), 3037 (vw), 3019 (vw), 2918 (s), 2894 (sh), 2846 (s), 2788 (sh), 2654 (vw), 2262 (w br), 1603 (s), 1587 (s), 1570 (m), 1537 (vw), 1505 (s), 1461 (m), 1446 (s), 1397 (s), 1352 (s), 1268 (m), 1215 (m), 1201 (m), 1185 (s), 1178 (s), 1159 (m), 1128 (s), 1117 (sh), 1105 (m), 1089 (sh), 1072 (w), 1039 (m), 1018 (sh), 998 (s), 951 (vw), 931 (m), 921 (m), 887 (m), 851 (s), 818 (m), 807 (sh), 770 (sh), 758 (m), 739 (s), 707 (w), 687 (s), 549 (w), 507 (s), 458 (m), 439 (m), 402 (w) cm1. NN-Bis(2-(di-tert-butylphosphino)ethyl)aniline (PNPhPtBu, 1 h) Following general procedure A, the PNPhPtBu (1h) ligand was obtained as white solid (1.39 g, 3.18 mmol, 79%). 31P{1H} NMR: (161.98 MHz, CDCl3, 300 K): δ=25.0 (s, 2 P, PtBu2) ppm. 1H NMR (400.13 MHz, CDCl3, 300 K): δ=7.17–7.06 (m, 2 H, CHphenyl), 6.62–6.52 (m, 3 H, CHphenyl), 3.44–3.39 (m, 4 H, N-CH2), 1.64–1.58 (m, 4 H, P-CH), 1.71–1.65 (m, 4 H, P-CH2), 1.08 (d, 3JPH =11.3 Hz, 36 H, CH3) ppm. 13C{1H} NMR (100.62 MHz, CDCl3, 300 K): δ=145.8 (s, 1 C, Ci,phenyl), 128.4 (s, 2 C, Co,phenyl), 114.6 (s, 1 C, Cp,phenyl), 110.7 (s, 2 C, Cm,phenyl), 50.6 (d, 2JPC =40.6 Hz, 2 C, N-CH2), 30.2 (d, 1JPC =19.4 Hz, 4 C, P-C) 28.6 (d, 2JPC =13.6 Hz, 12 C, CH3), 18.2 (d, JPC =22.7 Hz, 2 C, CH2-P) ppm. IR (ATR): ~ v=3093 (w), 3063 (w), 3039 (w), 3025 (w), 2982 (sh), 2950 (sh), 2937 (vs), 2892 (m), 2860 (m), 2706 (vw), 1910 (vw), 1750 (vw), 1602 (s), 1590 (w), 1569 (vw), 1541 (vw), 1503 (vs), 1462 (vs), 1452 (sh), 1408 (sh), 1393 (w), 1383 (s), 1358 (vs), 1344 (vs), 1321 (sh), 1280 (s), 1203 (sh), 1183 (vs), 1123 (vs), 1081 (vw), 1036 (s), 1014 (m), 992 (m), 933 (sh), 922 (m), 854 (m), 810 (s), 768 (sh), 750 (sh), 742 (vs), 707 (m), 688 (vs), 618 (vw), 593 (m), 578 (m), 545 (vw), 535 (vw), 506 (s), 464 (w), 434 (s) cm1. [Mo(CO)3(PNPhPPh)] (2 a) Following general procedure B, the [Mo(CO)3(PNPhPPh)] (2a) was obtained as yellow solid (45.2 mg, 0.07 mmol, 34%). Parts of the analytical data was already presented by Cowley et al.[26] 31P{1H} NMR: (161.98 MHz, CD2Cl2, 300 K): δ=38.0 (s, 2 P, PPh2) ppm. 1H NMR (400.13 MHz, CD2Cl2, 300 K): δ=7.78–7.73 (m, 4 H, CHphenyl), 7.49–7.46 (m, 2 H, CHphenyl), 7.42–7.35 (m, 6 H, CHphenyl), 7.33–7.28 (m, 2 H, CHphenyl), 7.27–7.20 (m, 2 H, CHphenyl), 7.19–7.14 (m, 4 H, CHphenyl), 7.13–7.07 (m, 3 H, CHphenyl), 6.99–6.96 (m, 4 H, CHphenyl), 3.71–3.59 (m, 2 H, N-CH2), 3.21–3.08 (m, 2 H, N-CH2), 2.85–2.69 (m, 4 H, P-CH2) ppm. 13C{1H} NMR (100.62 MHz, CD2Cl2, 300 K): δ=231.6 (t, 2JPC =8.24 Hz, 1 C, COaxial), 219.7 (ABX, 2 C, COequatorial), 153.8 (s, 1 C, N-(Ci,phenyl)), 139.1–138.5 (m, 2 C, P-(Ci,phenyl)), 137.0–136.7 (m, 2 C, P-(Ci,phenyl)), 132.4 (t, JPC =6.13 Hz, 4 C, P-(Co,phenyl)), 131.2 (t, JPC =6.33 Hz, 4 C, P- (Co,phenyl)), 130.0 (s, 2 C, P-(Cp,phenyl)), 129.4 (s, 2 C, P-(Cp,phenyl)), 129.1– 128.9 (m, 10 C, P-(Cm,phenyl), N-(Co,phenyl)), 125.3 (s, 1 C, N-(Cp,phenyl)), 121.1 (s, 2 C, P-(Cm,phenyl)), 58.0 (t, JPC =5.86 Hz, 2 C, N-(CH2)), 28.8 (dd, JPC =9.51 Hz, JPC =6.67 Hz, 2 C, P-CH2) ppm. IR (ATR): ~ v=3837 (vw), 3717 (vw), 3616 (vw), 3599 (vw), 3548 (vw), 3078 (vw), 3055 (w), 3017 (w), 3002 (vw), 2977 (vw), 2927 (vw), 2895 (w), 2874 (w), 2849 (w), 1921 (vs, CO stretch), 1802 (sh, br, CO stretch), 1775 (sh, CO stretch), 1600 (m), 1586 (m), 1572 (m), 1494 (m), 1480 (m), 1432 (s), 1424 (sh), 1407 (w), 1377 (vw), 1331 (w), 1310 (sh), 1299 (w), 1223 (m), 1185 (m), 1095 (s), 1068 (m), 1037 (m), 1029 (m), 998 (sh), 991 (w), 971 (vw), 944 (w), 920 (vw), 905 (m), 898 (sh), 852 (sh), 843 (w), 813 (sh), 804 (s), 763 (m), 742 (vs), 695 (vs), 671 (m), 651 (s), 634 (s), 610 (m), 597 (m), 554 (m), 530 (m), 515 (vs), 499 (vs), 487 (sh), 467 (m), 459 (sh), 449 (w), 428 (vs) cm1. Raman:~ v=3069 (vw), 3143 (vw), 3081 (w), 3056 (vs), 3007 (br, w), 2989 (vw), 2979 (vw), 2957 (w), 2950 (w), 2926 (w), 2916 (m), 2896 (vw), 2876 (vw), 2850 (vw), 2734 (vw), 2534 (vw), 1921 (w, CO stretch), 1828 (m, CO stretch), 1819 (vs, CO stretch), 1808 (sh, CO stretch), 1804 (s, CO stretch) 1601 (w), 1586 (vs), 1573 (m), 1496 (vw), 1485 (vw), 1466 (w), 1443 (sh), 1436 (vw), 1423 (vw), 1411 (vw), 1380 (vw), 1205 (w), 1188 (w), 1163 (w), 1141 (w), 1097 (w), 1041 (sh), 1028 (m), 999 (vs), 816 (vw), 804 (vw), 786 (w), 756 (vw), 715 (sh), 703 (vw), 691 (w), 676 (w), 652 (w), 637 (vw), 617 (m), 599 (vw), 555 (w), 531 (vw), 520 (w), 489 (m), 477 (sh), 469 (m), 454 (m), 430 (w), 422 (sh), 395 (vw), 375 (vw), 344 (vw), 325 (vw), 294 (vw), 269 (w), 255 (vw), 224 (w), 201 (vw), 184 (m), 157 (m), 138 (sh) cm1. [Mo(CO)3(PNPhPEt)] (2 c) Following general procedure B, the [Mo(CO)3(PNPhPEt)] (2c) was obtained as beige solid (122 mg, 0.24 mmol, 44%). Crystals suitable for x-ray analysis were obtained by slow evaporation of the complex in dichloromethane. Anal. Calcd C21H33P2O3NMo (505.4 g/mol): C 49.9, H 6.58, N 2.77; C 50.2, H 6.59, N 3.10. Wiley VCH Montag, 28.08.2023 2399 / 317727 [S. 11/15] 1 Eur. J. Inorg. Chem. 2023, e202300280 (11 of 14) © 2023 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202300280 10990682c, 0, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202300280 by Cochrane Germany, Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 61 Chapter 5 Project 3 - Surface functionalization of tridentate PNP and PPP Ligands of the equatorial carbonyl ligands in the IRRA spectrum as the dipoles are not parallel to the surface. As expected, the tricarbonyl complex bearing the tripod ligand showed global desactivation of the carbonyl ligands, due to the electron withdrawing properties of the gold surface. [262] Just as reported by Schlimm and Stucke the desactivation of the symmetrical A(1) stretch was partially compensated. This was observed even though the ligand does not have a conjugated π-system.[261,262] Fig. 5.3: Illustration of the molybdenum tricarbonyl complex bearing a modified tripodal P 3 ligand on a Au(111) surface. The complex is attached to a TATA-platform which is used to vertically deposit the system on the surface.[262] Besides pincer and tripodal systems that were reported by our group previously, [261,262] this project focuses around the last class of ligands with three donors. Aim of this project was the functionalization of tridentate PEP (E = N, P) ligands for surface chemistry. As in prior projects, the ligands are to be coordinated to Mo(0) tricarbonyl complexes as the stability of the complexes has been investigated in depth and no coligands are needed. A further goal is the electrochemical investigation and a possible suitability as a heterogeneous catalyst. The ligand chosen as tridentate counterpart to pincer systems was the PNPhPPh ligand that was part of the second project as well (cf. Project 2, Chapter 4). Besides the PNP system a PPP system analogous to the tripodal PPP system was designed. As a starting point for the purely phosphine based system, the tridentate prPP H P ligand was used, which is used by our group for the synthesis of the pentaPod ligand (cf. Project 1, Chapter 3). The general composition of the systems is modified to fit the above mentioned criteria: the anchor group, the spacer and the head group. In order to attach the ligands to the gold surface, a substrate specific anchor group is needed. For this, two different approaches were chosen. One approach 68 includes the introduction of an acetylene group which could be either attached to a platform as presented in previous work or clicked to an azide function forming a triazole ring. The azide needed for this cycloaddition is bound to the gold surface via an alkyl SAM with a thiol group capable of forming covalent bonds to the gold surface (Figure 5.4 top right). Besides the formation of a triazole ring by Huisgen’s cycloaddition [263] and a platform approach (Figure 5.4 top left), the ligands could be modified with a group capable of binding to a substrate. One approach to this is the utilization a thiol function as anchor group. The thiol group is part of a thiophenyl group included in the ligand backbone. The phenyl group now acts as the spacer group between the surface and the complex (Section 5.2, Figure 5.4 right). Within the scope of this thesis, this approach was only applied to a PNP type ligand, and is the main focus of this project. Fig. 5.4: Illustration of the approaches for depositing a Mo(0) tricarbonyl complex bearing tridentate PNP ligand on gold surface targeted in this project. 69 Chapter 5 Project 3 - Surface functionalization of tridentate PNP and PPP Ligands 5.1 Acetylene functionalization of the PNPhPPh ligand As mentioned above, the PNP ligand chosen for the modifications towards surface attachable systems was the PN Ph P Ph ligand. From the study in the second project (cf. Project 2, Chapter 4) the limitations of the PNP systems and their stabilities are known, which led to the decision that diphenylphosphine donors would be a well suited choice. Not only are they known to be good donors regarding molybdenum tricarbonyl complexes, but are also the least sensitive towards oxidation from the list of phosphines available. As known from previous experiences in our group, the systems will be exposed to some oxygen during preparation for the surface active spectroscopies. Either from transfer into the sample chamber or from oxygen/water adsorbed on the surface of the waiver. NOH HO NClCl NClCl Br 1. POCl3 reflux, 6 h 2. H2O [Bu4N][Br3] dcm, 0 °C CuI, [Pd(PPh3)4] TMSA, NEt3 115 °C, 3 d NClCl Si NPP Si n-BuLi, thf, 0 °C 2 eq. HPPh2 1 2 3 4 Scheme 1: Synthetic access to the surface modified PN Ph P Ph ligand 4.TMSA = Trimethylsilylacetylene. Starting from N-phenyldiethanolamine chlorination of the hydroxy groups gave N,N-bis(2chloroethyl)aniline (1). This was done by a modified method based on the procedure by Hrishikesan et al. [264] After the reaction was completed, excess phosphorus oxychloride was carefully neutralized with water. Cooling the reaction mixture using an ice bath suppresses 70 5.1 Acetylene functionalization of the PNPhPPh ligand the neutralization, leading to a spontaneous and rather violent reaction upon warming to room temperature, why a water bath at room temperature was used as coolant. This synthesis was also attempted using thionyl chloride, however no product could be obtained. The completeness of the conversion from the alcohol to the chloride (1) was confirmed by NMR and IR spectroscopy. Especially the infrared spectrum showed the successful conversion as no characteristic OH stretches were found. In order to introduce the acetylene function, the phenyl ring on the aniline was brominated in the 4-position, yielding 4-bromo-N,N-bis(2-chloroethyl)aniline (2). This was achieved by following the procedure described by Huffman et al., [265] however the reaction was cooled to 0°C instead of 10 °C. The crude product was dissolved in methanol and crystallized in the freezer (-32 °C). The solid was filtered and solved in methanol again to decrease the amount of [n-Bu 4 N] + salts from the product. In order to fully remove the salts the crude product was recrystallized from methanol and the desired compound 2was obtained as light pink solid. Introduction of the acetylene group was achived by Sonogashira coupling [266] of compound 2to obtain N,N-bis(2-chloroethyl)-4-((trimethylsilyl)ethynyl)aniline (3). Multiple attempts were made in order to obtain the desired product. The first attempts had a reaction time of 5 d, which was later reduced to 3 d. Futhermore, the catalyst was varied and changed from [ PdCl2(PPh3)2 ] to [ Pd(PPh3)4 ]. [266,267] After column chromatography using n-hexane and ethyl acetate (4:1) the precursor to ligand 4was obtained as dark orange oil. From the NMR spectra it became evident that, although purified by column chromatography, small impurities remained. Nonetheless, the introduction of the TMS-acetylene group was observed in the 1 Hand 13 C NMR spectra. Despite small impurities the precursor to the ligand was further reacted using the well established route for the introduction of the secondary phosphines with n-butyllithium and subsequent purification by filtration over Celite ® and basic aluminum oxide. [216–218,268] This yielded the desired product N,N-bis(2-(diphenylphosphino)ethyl)-4- ((trimethylsilyl)ethynyl)aniline (4) as highly viscous orange oil (44%). 71 Chapter 5 Project 3 - Surface functionalization of tridentate PNP and PPP Ligands Fig. 5.5: 31 P {1 H } NMR spectrum of ligand 4. The successful substitution of the terminal chlorides by diphenyl phosphine can be seen from the product signal at -21.1 ppm. From the 31 P NMR spectrum (Figure 5.5) the formation of ligand 4was confirmed. This is indicated by the signal at -21.1 ppm, which is known from the unmodified ligand. Beside the main product signal, those of some byproducts were found as well. The signal at -40.0 ppm can be assigned to excess diphenyl phosphine. Three other small impurities were found at -23.0 ppm, -20.9 ppm and -14.5 ppm which could not be identified but can possibly assigned to a singular substituted product as not all of the HPPh2 has reacted and potentially some decomposed product. The formation of a small byproduct is also confirmed from the 13 C NMR spectrum as a second, smaller set of signals is found beside the product signals of the ligand. The carbon atoms of the acetylene unit on the other hand are only found once. Considering the slight high field shift of the signals, the byproduct is either the phosphinated version of the brominated compound 2or the bromide as well as the chloride substituents were exchanged by the lithium diphenyl phosphide. As can be seen from the integrals of the signals in the 31 P NMR spectrum, the impurities are rather small, which is why the ligand was still been used for the coordination to molybdenum tricarbonyl precursors. The first attempts were made using [Mo(CO) 3 ( η6 -toluene)] as precursor complex, due to the [Mo (CO)3 (cht)] complex not being available at that time. The [Mo(CO) 3 ( η6 -toluene)] complex was synthesized from [ Mo(CO)6 ] after modified procedures 72 5.1 Acetylene functionalization of the PNPhPPh ligand from Rybinskaya et al. [269] and Whiting &Nicholls. [270] The [Mo(CO) 3 ( η6 -toluene)] was chosen as the complex is less stable compared to the cycloheptatriene analogue and more prone towards ligand exchange. [271] Nonetheless, no reaction was observed at room temperature. Due to this, a second attempt with increased temperature was made. This lead to the formation of a precipitate, which was filtered and washed. The IR spectrum (Figure 5.6, black spectrum) of the obtained complex showed a multitude of CO stretches, indicating a mixture of complexes. Although the stretches of the desired complexes are visible as shoulders, a purification of the mixture was not successful and the attempt was discarded. Fig. 5.6: Comparison of the infrared spectra of the [Mo (CO)3 (PN Ph P Ph )] complex (top, red) to the attempted coordination of the surface modified PNP ligand 4(bottom, black). During the experimental work of Project 2, some experiments were conducted using [Mo(CO) 3 ( η6 - toluene)]. Nonetheless, the best results were achieved using the cycloheptatriene molybdenum tricarbonyl complex and as it was not commercially available at the time, it was synthesized from Mo(CO)6 . The first approach to the synthesis was carried out by following a slightly modified procedure of Bisnette et al. [272] Refluxing Mo(CO)6 in dry n-octane and cyclohepatriene (cht) gave a dark red solution. Removal of the solvent and cht has to be performed with extrem caution and the temperature should not exceed 35 °C as the complex would sublimate. The dried crude product was then transferred into a Soxhlet extractor and extracted using n-hexane. The solvent was removed carefully and the complex obtained as red solid. Due to the poor yield of the Soxhlet extraction, an alternative procedure was used for the isolation and purification of the complex during a second run. After reflux and 73 Chapter 5 Project 3 - Surface functionalization of tridentate PNP and PPP Ligands removal of the volatile components, the [Mo (CO)3 (cht)] complex was extracted from the crude product upon addition of n-hexane. This step was repeated multiple times until the solvent did not dyed red anymore. This step removed some parts of unreacted Mo(CO)6 , the rest was removed via sublimation. The temperature applied should not exceed 30 °C to assure the sublimated compound is the hexacarbonyl only. This process is very time consuming but yields a very pure product. Unfortunately, both procedures resulted very poor yields. The poor yield can be attributed to the Mo(CO)6 sublimating during the reflux in the first step of the reaction. Clausen was able to improve the yield by changing the apparatus and building a custom spatula and scraping the sublimated Mo(CO)6 back into the solution. In addition to that the reaction time was significantly elongated.[273] Using the [Mo (CO)3 (cht)] complex, another attempt to coordinate the surface modified PN Ph P Ph ligand (4) was made. A successful coordination of the PNP ligands to the Mo(CO) 3 fragment should result in the precipitation of a yellowish/beige solid. Using identical conditions for the coordination as used for the unmodified ligand, the solution remained unchanged. Due to this, the temperature was increased to 50 °C for 6 h and afterwards stirred for another 12 h. This resulted in a darkening of the solution but not to the formation of the anticipated yellowish solid, which was observed for nearly all molybdenum tricarbonyl complexes bearing PNP ligands in Project 2. Nonetheless, the solvent was removed and the crude product washed multiple times with n-pentane to remove ligand and precursor complex. The remaining solid was washed and dried. The synthesis resulted again in a mixture of complexes. Although the desired complex was contained in the product mixture, multiple side reactions occurred as can be seen from 31P NMR spectrum (Figure 5.7) 74 5.1 Acetylene functionalization of the PNPhPPh ligand Fig. 5.7: 31 P {1 H } NMR spectrum of the attempted coordination of the surface modified PN Ph P Ph ligand 4using [Mo(CO)3(cht)]. From comparison to the spectrum of the unmodified PN Ph P Ph ligand it can be seen that the desired complex was formed but many other species as well. By comparing with the literature, the signal at 25 ppm can be assigned to the tetracarbonyl complex with the phosphines coordinated. [274] The tetracarbonyl complex is the product of a decomposition reaction of the desired complex. This can possibly be retraced to the increased temperatures applied. The desired product is expected at around 40 ppm. As can be seen from Figure 5.7, multiple species were found, indicating the formation of multiple tricarbonyl complexes with facial geometry. This leads to the conclusion that the ligand decomposed to a certain degree as well. Unfortunately, the desired carbonyl complex could not be isolated but should be obtainable under different conditions e.g. elongation of reaction times, different solvents and lower temperatures. Although the attempts made are very promising and ligand 4was synthesized successfully, the main focus was redirected to another variant of the PN Ph P Ph ligand that features a thio-based anchor group. 75 Chapter 5 Project 3 - Surface functionalization of tridentate PNP and PPP Ligands 5.2 Thiofunctionalization of the PNPhPPh ligand The second variant of a surface functionalized PNP ligand features a thiol anchor group. This alternative approach is another attempt continuing on from the result presented in the second project of this thesis. The sulfur anchor is chosen in order to create a direct covalent bond between the system and the gold surface. Sulfur is an excellent choice when the surface aimed for is gold as the aurophillic character of the sulfur helps with the formation of stable bonds. [275] The first steps of the reaction are similar to the ones of the acetylene functionalized PNP ligand. N SCN Cl Cl N SCN PPh2 Ph2Pn-BuLi 0°C, thf 2 eq. HPPh2 1. Br2, NaBr 2. NH4SCN NClCl NOHHO reflux, 6 h POCl3 dry MeOH [Mo(CO)3(cht)] toluene, roomtemp. Et2O, 0°C LiAlH4 Mo Ph2P Ph2PCO CO CO N SCN Mo Ph2P Ph2PCO CO CO N SH 1 5 6 7 8 Scheme 2: Synthetic access to thiocyanate-functionalized PN Ph P Ph ligand (6) and the corresponding tricarbonyl complex 7. Herein the SCN group acts as protective group during the coordination and is converted to a surface active thiol via reduction. The first step is a chlorination following the procedure described in the sections 4.1 and 5.1 resulting the PNP precursor 1. The sulfur anchor is introduced as thiocyanate which was chosen for multiple reasons. The first reason is that the introduction of a thiocyanate group 76 5.2 Thiofunctionalization of the PNPhPPh ligand is a well established synthesis and is selective towards the para-position of the amine. The second reason is that the thiocyanate group enables two methods of attachment to the gold surface. The first involves reduction to a thiol and subsequent binding to the surface (scheme 2) while the second method is a direct attachment upon a gold induced CN - release. [276] The introduction of the SCN group is a two step reaction (Scheme 3). In the first step of the reaction a bromine is introduced in the 4-position. This can be done in two ways: either by following the Huffman procedure [265] described in section 5.1 using [NBu 4 ][Br 3 ] or following the instructions of Creighton et al. generating the reactive Br 3ion in situ from bromine and sodium bromine in dry methanol and adding it to a solution of 1and ammonium thiocyanate. The first variant works best when isolating the 4-bromo-N,N-bis(2-chloroethyl)aniline (2), dissolving it in methanol and adding pure NH4SCN . The product can be recrystallized from ethanol. The advantage of the second method is that it is a one-step synthesis. The crude product of the synthesis is a brown oil, which is best purified by recrystallization from ethanol. Both methods were tried, both leading to a successful isolation of the desired product in good yields. N Cl Cl N Cl Cl Br N Cl Cl SCN NH4SCN MeOH [NBu4][Br3] dcm 1. Br2, NaBr 2. NH4SCN MeOH 125 Scheme 3: Illustration of the two methods used to introduce the thiocyanate group into the ligand precursor. N,N-bis(2-chloroethyl)-4-thiocyanatoaniline (5) is an ideal dummy molecule in order investigate and optimize the direct deposition upon CN - release on a gold substrate. The main disadvantage is that investigations with IRRAS are problematic as no good indicator regarding the surface orientation is present in the molecule. Nonetheless, first attempts regarding a direct deposition have been made. Although the preparation method still requires optimization, it was found that a first attempt of the direct deposition worked to some extent as can be seen from the XP spectra (Figure 5.8). 77 Chapter 5 Project 3 - Surface functionalization of tridentate PNP and PPP Ligands Fig. 5.10: Comparison of the aromatic region of the 13 C {1 H } NMR spectra of the PN Ph P Ph ligand (top) and the borane protected PN Ph P Ph after the introduction of the thiocyanate group (bottom). The introduction of the SCN group leads to a shift of the two meta carbons (blue arrow) and the para-carbon (red arrow) of the aniline ring. The 13 C NMR spectrum shows the successful introduction of the SCN group into the PN Ph P Ph giving the ligand precursor 11. Furthermore, the 31 P NMR spectrum shows that the halogenation of the phosphines is only observed as minor side reaction and the borane protection has withstood the reaction conditions, especially the release of Br2(Figure 5.11, bottom). 84 5.2 Thiofunctionalization of the PNPhPPh ligand Fig. 5.11: Comparison of the 31 P {1 H } NMR spectra of the PN Ph P Ph ligand (9, top), the borane protected PN Ph P Ph ligand (10, middle) and the borane protected PN Ph P Ph after the introduction of the thiocyanate group (11, bottom). The borane groups successfully withstood the reaction conditions during the introduction of the SCN group. Besides the main product one small, broad signal is visible as well as two small, low field shifted signals. The final step in the ligand synthesis is the removal of the borane groups. The common way to remove them is by use of morpholine and temperature. The ligand precursor 11 was refluxed in morpholine and subsequently stirred at 90 °C for 3 d. The conditions applied unfortunately lead to the formation of a mixture of products as can be seen from the 31 P NMR spectrum (Figure 5.12). The products formed can be assigned to different species. The desired product 6, a small fraction of the ligand with one brominated phosphine group and one non-halogenated phosphine. Furthermore, four additional signals appeared between 30.6 and 30.3 ppm. These signals may be associated with phosphine oxide variants formed during the attempted removal of the protective group. The signals found could be assigned to different combinations of fully oxidized phosphines, with only one phosphine oxide and a minor combination of an oxidized and a halogenated phosphine. This 85 Chapter 5 Project 3 - Surface functionalization of tridentate PNP and PPP Ligands assumption is based on the 31 P NMR shifts reported for different molecules based around functionalized ethyldiphenylphosphine oxides. [283] The formation of the oxides indicates that, although Schlenk conditions were applied, the phosphines must have been exposed to oxygen. This is probably due to insufficient degassing of the morpholine prior to use. Fig. 5.12: 31 P {1 H } NMR spectrum of the attempted removal of the borane groups on SCN modified PN Ph P Ph ligand (11). Besides the formation of the product and the halogenated phosphines from the prior step mainly phosphine oxides were formed. Although the ligand variant with phosphine oxides would not coordinate, no coordination attempt was performed, as only a very small part of the product obtained can be assigned to the desired PN Ph-SCN P Ph ligand (6). Nonetheless, this route is very promising and should be repeated with some variations of the reaction conditions and thoroughly degassed morpholine. Besides the use of morpholine as removal agent for the protective groups, a different amine like triethylamine could be tried. 86 5.3 Functionalization of the prPPHP-ligand 5.3 Functionalization of the prPPHP-ligand Besides the effords sourrounding the surface modifications of the PNP ligand, attempts were made to modify a well known building block from the pentaPod synthesis - the prPPHP. The ligand was supposed to represent the tridentate PPP counterpart to the already successful deposited tripodal P3 system by Petersen et al. [262] The underlying synthetic route is based on a modular system (Scheme 6). The synthesis of the prPPHP ligand [145,231] is well established in our group and halogen-containing building blocks that not only have the appropriate anchor group, but also the desired spacer are commercially available. In line with the projects already presented, the tridentate PPP system was also to be equipped with an acetylene unit to bind to a platform or an alkyl-SAM for surface fixation. n-BuLi 0°C, thf P H Ph2PPPh2 PPh2PPPh2 TMS X TMS + X = F, Cl, Br 12 13 Scheme 6: Reaction scheme for the modular assembly of the surface-modified prPPHP ligand (13) based on the prPPHP ligand (12) and the ((4-halogenephenyl)ethynyl)trimethylsilane. The idea behind this route was to apply known substitution of halogens by phosphines. Following the well established route the prPPHP ligand was synthesized. In a first attempt the prPPHP ligand was dissolved in thf and n-BuLi was added. Parallel to that, the ((4chlorophenyl)ethynyl)trimethylsilane was dissolved in thf and subsequently added dropwise to the prPPHP solution. The result of this reaction was a mixture of multiple species. As the terminal phenyl phosphines are quite resistant to air an attempt was made to purify the desired product using column chromatography. From the 31 P NMR spectrum of the isolated product it can be seen that the two building blocks were connected successfully (Figure 5.13, bottom). 87 Chapter 5 Project 3 - Surface functionalization of tridentate PNP and PPP Ligands Fig. 5.13: Comparison of the 31 P {1 H } NMR spectrum of the prPPHP ligand (top) and the modified ligand 13 (bottom). The enlarged regions show several byproducts and oxidized species. The large signal at -17.0 ppm can be assigned to the PPh 2 groups and the signal at -26.1 ppm to the central phosphine after the substitution of the proton by the aromatic residue. As can be seen from the top spectrum, the central PH of the prPPHP ligand can be found at -72.7 ppm. After the reaction the signal of the central phosphorus shifted to -26.1 ppm due the substitution of the hydrogen by the aromatic residue. Nonetheless, multiple by-products were still found in the product mixture. Since a second attempt at column chromatographic purification was deemed unlikely to be successful, due to repeated exposure to oxygen, a coordination attempt was started despite the lack of purity in the hope that this would yield a pure product. The impure ligand was added to the [Mo (CO)3 (cht)] precursor complex. As it was already observed for the tricarbonyl complexes with PNP ligands, the coordination was 88 5.3 Functionalization of the prPPHP-ligand indicated by a color change from deep red to yellow/beige. Although the yield was poor, a solid was isolated. For analysis via NMR spectroscopy, the solid was dissolved in DCM, which led to the formation of an insoluble black solid and thus to the presumed decomposition of the product. Furthermore, an IR spectrum of the isolated product was measured. Although the color of the solid looked promising, the spectrum showed that the desired tricarbonyl complex was not obtained, because no vibrations were found in the CO stretching frequency region. By modifying the synthesis parameters and substituting the chlorine-based building block with a brominated and flourinated analogue, the objective was to synthesize the pure ligand. The synthesis was carried out under the same reaction conditions as in the previous experiment. Unfortunately, the desired ligand was not obtained. In subsequent attempts the temperatures were varied and the n-BuLi was substituted by lithium diisopropylamide which also did not provide the desired results. In parallel with efforts to synthesize the prPP Ph-acetylene-TMS P ligand (13), the successful synthesis of the ethyl-bridged analog was demonstrated by our group. [284] It was found that in order to obtain a pure product, a halogen atom is required as substituent on the central phosphine. Nonetheless, it was also demonstrated that the linking of the prPPHP ligand and the ((4-chlorophenyl)ethynyl)trimethylsilane is possible in principle. 89 Chapter 6 Conclusion & Outlook This thesis is split into three main projects. The first focuses around the different behavior of the [ W(N2) (P 2Me PP 2Ph )] complex compared to its molybdenum analogue. The investigations were based around the assumption that the complexes have different redox potentials, why the main focus evolved around electrochemistry. The second study provides an in-depth investigation regarding the coordination behavior of different tridentate PNP ligands. All PNP ligands were coordinated to molybdenum tricarbonyl complexes, due to their excellent spectroscopic properties. Gaining a better understanding of these systems is beneficial for the final project of this thesis, which deals with the synthesis and characterization of molybdenum tricarbonyl complexes bearing tridentate PNP ligands with functionalization for surface deposition. This project aims to continue the work of Stucke,Schlimm and Petersen et al., which already demonstrated the deposition of molybdenum tricarbonyl complexes with pincer and tripodal ligands on surfaces.[261,262] 6.1 Project 1 The first project was focused on synthetic nitrogen fixation with the pentaPod ligand and a continuation of a prior study by Engesser et al., which demonstrated the catalytic activity of the molybdenum dinitrogen complex bearing the pentaPod and its superiority compared to the regular dinitrogen complexes with pentaphosphine coordination spheres provided by triand bidentate phosphine ligands. As the molybdenum complex was catalytically active, the next step in testing the capability of the ligand was the coordination to alternative metal center. As tungsten is usually the metal of choice when exchanging molybdenum, the pentaPod was coordinated on tungsten and investigated regarding its activity towards derivatization and ammonia formation. This project was executed in collaboration with Junge and the pentaPod ligand was successfully coordinated and the tungsten mono-dinitrogen complex isolated. The resulting [W N2 (pentaPod)] complex was characterized using IR, Raman and NMR spectroscopy. Besides an expected shift to lower wavenumbers of the N2 -stretch and additional satellites in the 31 P-NMR spectrum, the spectra as well as the crystal structure were surprisingly similar to the molybdenum analogue. Investigations regarding the catalytic activity towards ammonia formation of the [ W(N2) (P 2Me PP 2Ph )] complex in the presence of samarium diiodide / water resulted in 2.75 equiv. Although the activity was found to be well below the results of the molybdenum complex, this is the first example of a tungsten 91 Chapter 6 Conclusion & Outlook dinitrogen complex capable of generating more than two equivalents of ammonia. In order to investigate the differences of the complexes, electrochemical and spectroelectrochemical studies were performed. The molybdenum as well as the tungsten complex were investigated by cyclic voltammetry. The cvs found for the complexes were very similar. E 1/2 of the first redox systems were determined at -1.16 V for the tungsten and at -1.13 V for the molybdenum complex, respectively. Nonetheless the reversibility of the 0 ⇄ +I system differs, as the tungsten complex was found to be fully reversible unlike the molybdenum complex, which was only partly reversible. For the second oxidation systems very similar E 1/2 values were found as well, yet the reversibility differed. Both systems were irreversible at low scan rates, but an increase in reversibility was observed for the tungsten complex as the scan rate was increased. Further analysis of the peak current function using Randles-Sevcik equation revealed an ECE type mechanism for the tungsten complex. Deeper investigations into the species formed upon oxidation were carried out by means of IR-spectroelectrochemical investigations. It was found that the [ W(N2) (P 2Me PP 2Ph )] + complex was more stable than the molybdenum analogue, which released the dinitrogen ligand upon oxidation. Due to the higher stability of the oxidized complex the tungsten-N2complex was observed. Following this project Junge et al. presented an extensive theoretical study regarding possible mechanisms of the NH3 formation and the competing hydrogen evolution reaction. [230] From previous attempts within the framework of this thesis, it was found that no significant amounts of ammonia were generated electrocatalytically upon addition of acids (cf. section 3.2). Further experimental investigations regarding this topic could be based on the success in electrocatalytic ammonia formation presented by the Peters group. [173] They demonstrated that the catalytic activity of a tungsten complex can be achieved by adding a mediator that acts as a PCET agent. 6.2 Project 2 The second and main project of this thesis was the investigation of the coordination behavior of tridentate PNP ligands on molybdenum tricarbonyl systems. The scope was to investigate the influence of phosphines and their substitutes on coordination behavior and stability. For this, a tridentate ligand with central aniline donor was chosen in reference to the well studied pincer ligands. A series of tridentate PN Ph P R ligands (R = Ph, Me, Et, Pln, Cyp, i Pr, Cy, t Bu) was synthesized and coordinated to the [Mo(CO) 3 (cht)] precursor. All ligands were successfully synthesized but not all complexes could be obtained. Complexes bearing the PN Ph P Me , PN Ph Pln and PN Ph P tBu could not be isolated. All other complexes were isolated and thoroughly characterized. From IRand Raman spectra it became apparent that all complexes adapted a facial geometry in the solid state. This was expected as the precursor 92 6.2 Project 2 predefined a facial coordination geometry. The facial geometry was also found in solution for all complexes, which was determined from the shift of the signals in the 31 P NMR spectra as well as the ABX pattern found for the equatorial CO ligands in the 13 C NMR spectra. Nonetheless, the meridional isomer was found as minor byproduct (<1%). Comparison to similar systems (NH instead of NPh) in the literature revealed that the geometry of the complexes shifted from a facial to a meridional geometry with increasing steric demand of the phosphines substituents. DFT studies were performed in order to gain insights into the isomerization mechanism and the discrepancies between the two systems. Isodesmic calculations of the relative ligand exchange energies revealed a contradiction to the experimental results for the systems containing NPh as the meridional isomer is energetically favored for bulky phosphines. For the systems with NH, the calculations agree well with the experimental observations. What became clear, however, is that the molybdenum tricarbonyl complex bearing the PNPhPtBu ligand was energetically unfavorable in the facial and meridional geometry. Nonetheless, in the calculations no salient difference crystallized between the NPh and NH systems that would explain the different behavior, excluding thermodynamics as the cause for the observed geometries. Further calculations regarding the kinetics were performed in order to investigate the isomerization mechanism. Two possible pathways were found. The first is a direct isomerization in which one isomer transforms into the other over a transition state. The transition state during direct isomerization proved to be energetically unfavorable and also showed no significant differences. The second pathway involves the dissociation of a CO ligand prior to the isomeric transformation. This pathway indeed revealed differences between the two systems. For the transformation of the NH system the transition state between the facial and meridional state (starting from the dicarbonyl variant) was found to be energetically lower compared to the NPh complex. Furthermore, a second, low lying transition state between two conformers of the meridional isomer was found for the NH system. Of these two only one belongs to the conversion of the facial and the meridional isomers into each other. The higher energies of the NPh system can be explained by the formation of an agostic hydrogen bond between the phenylring and the molybdenum, stabilizing the pentacoordinated species, which needs to be broken during the isomerization, therefore leading to an higher energy barrier, hence inhibiting the isomeric transformation. The experimental results of the complexes bearing PN Ph P ligands were obtained due to the fac-preorientation caused by the cyclohepatriene precursor. Subsequent investigations include the substitution of this precursor by a complex with prior meridional ligand arrangement or no predetermined orientation like the utilization of molybdenum hexacarbonyl as demonstrated by Beller et al. [285,286] Although high temperatures are required for this synthesis, Mo(CO) 6 does not dictate a preferred coordination geometry, which could allow for the formation of the meridional isomers for ligands with bulky phosphines. Another possibility would be to coordinate the ligands to a molybdenum dicarbonyl complex, generating the 93 Chapter 7 Experimental Section 13 C-NMR (125 MHz , CDCl3 , 300 K ): δ =146 (s, 1 C, C-3), 134 (s, 2 C, C-4), 114 (s, 2 C, C-5), 111 (s, 1 C, C-6 ), 106 (s, 1 C, C-10 ), 92.2 (s, 1 C, C-9), 53.4 (s, 2 C, C-1), 40.4 (s, 2 C, C-2), 0.30 (s, 3 C, C-11) ppm. 7.4 N,N-Bis(2-(diphenylphosphanyl)ethyl)-4- ((trimethylsilyl)ethynyl)aniline NClCl Si NP P Si n-BuLi, thf, 0 °C 2 eq. HPPh2 3 4 N,N-bis(2-chloroethyl)-4-((trimethylsilyl)ethinyl)aniline (3, 440 mg, 1.34 mmol) was dissolved in thf (10 mL). In a second vessel, diphenyl phosphine (521 mg, 2.80 mmol) was dissolved in thf (10 mL) as well and both solutions cooled to 0 °C. To the diphenyl phosphine solution, n-buli (1.2 mL, 3.14 mmol) was added dropwise and stirred for 30 min. Afterwards, the solution with the ligand precursor 3was added slowly to the phosphine solution and stirred for 4 d at room temperature. The solvent as well as excess n-buli were removed in vacuo. The residue was dissolved in diethyl ether and filtered through basic aluminum oxid and Celite ® . The product was obtained as viscous orange oil. Yield: 360 mg (0.59 mmol) 44 % 100 7.4 N,N-Bis(2-(diphenylphosphanyl)ethyl)-4-((trimethylsilyl)ethynyl)aniline 1 2 3 4 567 8 9 NP P Si 10 11 12 13 1 H-NMR (400 MHz , CDCl3 , 300 K ): δ =7.31-7.26 (m, 8 H, H-11), 7.22-7.16 (m, 12 H, H-12/H-13), 7.09-7.05 (m, 2 H, H-5), 7.02-7.00 (m, 2 H, H-4), 3.25-3.17 (m, 4 H, H-2), 2.17-2.12 (m, 4 H, H-1), 0.12 (m, 9 H, H-9) ppm. 13 C-NMR (100 MHz , CDCl3 , 300 K ): δ =147.7 (s, 1 C, C-3), 138.8 (d, J CP = 12.3 Hz, 4 C, C-10), 133.3 (s, 2 C, C-4), 132.8 (d, J CP = 18.9 Hz, 8 C, C-11), 128.9 (s, 4 C, C-13), 128.6 (d, J CP = 6.76 Hz, 8 C, C-12), 111.5 (s, 2C, C-5), 109.8 (s, 1 C, C-6), 106.5 (s, 1 C, C-7), 91.3 (s, 1 C, C-8), 47.7 (d, J CP = 25.6 Hz, 2 C, C-2), 26.2 (d, J CP = 14.4 Hz, 2 C, C-1), 0.37 (s, 3 C, C-9) ppm. 31P-NMR (162 MHz, CDCl3, 300 K, H3PO4): -21.9 (s, 2P, PPh2) ppm. 101 Chapter 7 Experimental Section 7.5 N,N-Bis(2-chloroethyl)-4-thiocyanatoaniline N SCN ClCl 1. Br2, NaBr 2. NH4SCN NCl Cl dry MeOH 1 5 N,N-Bis(2-chloroethyl)aniline (1, 2.00 g, 9.17 mmol) and ammonium thiocyanate (1.5 g, 19.7 mmol) were dissolved in methanol (30 mL). The solution was cooled to 0 °C and a second solution of bromine (0.5 mL) and sodium bromide (0.5 g) dissolved in methanol (5 mL), was added drop wise. The solution was stirred for 10 min at 0 °C, before poured into deionized water (100 mL). This step was supposed to precipitate the desired product, however, the crude product immediately dissolved again. Due to this, the solution was extracted with diethyl ether (3x 100 mL) and washed with sodium thiosulfate to remove excess bromine. The organic layer was dried over anhydrous magnesium sulfate and the solvent removed. The resulting dark oil was recrystallized from ethanol yielding the desired product as white solid. Yield: 1.39 (5.05 mmol) 55 % N Cl Cl N Cl Cl SCN TBABr3, NH4SCN DCM, MeOH, 0 °C 1 5 A second method was used in order to synthesize the N,N -Bis(2-chloroethyl)-4-thiocyanatoaniline (5). For this, N,N-Bis(2-chloroethyl)aniline (1, 2.00 g, 9.17 mmol) and ammonium thiocyanate (1.5 g, 19.7 mmol) were dissolved in a mixture of methanol (30 ml) and dichloromethane (10 mL). The solution was cooled to 0 °C and a solution of TBABr 3 (4.42 g, 9.17 mmol) dissolved in DCM (10 mL) was added. The solution was stirred for 10 min at 0 °C and another 30 min at room temperature. The solution was washed with sodium thiosulfate solution and dried using magnesium sulfate. The solvent was removed and the crude product recrystallized 102 7.6 N,N-Bis(2-(diphenylphosphino)ethyl)aniline-borane-complex from methanol. The recrystallization was repeated twice in order to remove all ammonium salts. Yield: 1.44 g (5.23 mmol) 57 % 1 2 3 4 56 7 N Cl Cl SCN 1 H-NMR (400 MHz , CDCl3 , 300 K ): δ =7.48-7.44 (m, 2 H, H-4), 6.72-6.68 (m, 2 H, H-5), 3.79-3.75 (m, 4 H, H-2), 3.66-3.62 (m, 4 H, H-1) ppm. 13 C-NMR (125 MHz , CDCl3 , 300 K ): δ =148 (s, 1 C, C-3), 135 (s, 2 C, C-4), 113 (s, 2 C, C-5), 112 (s, 1 C, C-6), 112 (s, 1 C, C-7), 53.4 (s, 2 C, C-2), 40.2 (s, 2 C, C-1) ppm. 7.6 N,N-Bis(2-(diphenylphosphino)ethyl)aniline-borane-complex BH3 SMe2 N Ph2P PPh2N Ph2P PPh2 BH3BH3 MeOH 910 The PN Ph P Ph (9, 604 mg, 1.17 mmol) was dissolved in DCM (15 mL) and the borane dimethyl sulfide complex (340 mg, 4.47 mmol) was added. The solution was stirred for 2 h at room temperature and all volatile compounds removed in vacuo. The product was obtained as colorless solid. Yield: 578 mg (1.06 mmol) 91 % 103 Chapter 7 Experimental Section 1 2 3 4 56 78 9 NP P 10 11 BH3 BH3 1 H-NMR (400 MHz , CDCl3 , 300 K ): δ =7.69-7.64 (m, 8 H, H-8), 7.60-7.58 (m, 2 H, H-4), 7.53-7.41 (m, 12 H, H-9/H-10), 7.39-7.34 (m, 3 H, H-5/H-6), 3.47-3.38 (m, 4 H, H-2), 1.85-1.74 (m, 4 H, H-1), 1.24-0.50 (m, 6 H, H-11) ppm. 13 C-NMR (100 MHz , CDCl3 , 300 K ): δ =147.7 (s, 4 C, C-7), 145.7 (s, 1 C, C-3), 131.2130.7 (m, 8 C, C-8), 128.5 (s, 2 C, C-4), 128.1-127.8 (m, 8 C, C-9 ), 127.4 (s, 2 C, C-5 ), 123.2 (s, 4 C, C-10), 111.7 (s, 1 C, C-6), 60.7 (d, J CP = 9.27 Hz, 2 C, C-2), 20.1 (d, J CP = 38.1 Hz, 2 C, C-1) ppm. 31P-NMR (162 MHz, CDCl3, 300 K, H3PO4): 12.5 (s, 2P, PPh2·BH3) ppm. 11B-NMR (128 MHz, CDCl3, 300 K, BF3): -41.3 (s, 2 B, BH3) ppm. 7.7 N,N-Bis(2-(diphenylphosphaneyl)ethyl)-4-thiocyanatoanilineborane-complex DCM, 0 °C N Ph2PPPh2 BH3BH3 N Ph2PPPh2 BH3BH3 SCN NH4SCN TBABr3 10 11 104 7.7 N,N-Bis(2-(diphenylphosphaneyl)ethyl)-4-thiocyanatoaniline-borane-complex PN Ph P Ph ·BH3 (10, 200 mg, 0.37 mmol) and ammonium thiocyanate (56.3 mg, 0.74) were dissolved in a mixture of DCM (5 mL) and methanol (5 mL). In a second vessel TBABr 3 (177 mg, 0.37 mmol) was dissolved using the same mixture. The ligand solution was cooled to 0°C and the TBABr 3 solution was added, changing the color to a deep purple color. After stirring for 10 min, sodium thiocyanate solution (5 drops) were added. All volatile components were removed in vacuo and the crude product isolated using diethyl ether. The product was obtained as colorless solid. Yield: 84.2 mg (0.14 mmol) 37 % 1 2 3 4 56 7 89 NP P 10 11 BH3 BH3 SCN 12 1 H-NMR (400 MHz , CDCl3 , 300 K ): δ =7.68-7.61 (m, 8 H, H-9), 7.55-7.34 (m, 12 H, H-10/H-11), 7.34-7.32 (m, 2 H, H-4), 7.19-7.15 (m, 2 H, H-5), 3.45-3.39 (m, 4 H, H-2), 2.44-2.36 (m, 4 H, H-1), 1.73-0.82 (m, 6 H, H-12) ppm. 13 C-NMR (100 MHz , CDCl3 , 300 K ): δ =147.7 (s, 1 C, C-3), 135.0 (s, 4 C, C-8), 132.2 (m, 2 C, C-4), 132.1 (s, 8 C, C-9), 131.8 (m, 8 C, C-10), 129.6 (s, 2 C, C-5), 129.2 (s, 4 C, C-11), 129.6 (s, 1 C, C-6), 108.3 (s, 1 C, C-7), 45.3 (d, J CP = 5.80 Hz, 2 C, C-2), 23.4 (d, JCP = 33.4 Hz, 2 C, C-1) ppm. 31P-NMR (162 MHz, CDCl3, 300 K, H3PO4): 12.5 (s, 2P, PPh2·BH3) ppm. 11B-NMR (128 MHz, CDCl3, 300 K, BF3): -41.3 (s, 2 B, BH3) ppm. 105 Chapter 8 References [1] S. Tong, H. Bambrick, P. J. Beggs, L. Chen, Y. Hu, W. Ma, W. Steffen, J. Tan, Environ. 2022,158, 106892. [2] P. Dasgupta, A. Dasgupta, S. Barrett, Environ Resource Econ 2021. [3] J. Tyagi, S. Ahmad, M. Malik, Int. J. Environ. Sci. Technol. 2022. [4] W. M. Stewart, D. W. Dibb, A. E. Johnston, T. J. Smyth, Agron. J. 2005,97, 1–6. [5] V. Smil, Ambio 2002,31, 126–131. [6] J. Erisman, M. Sutton, J. Galloway, et al., Nature Geosci 2008,1, 636–639. [7] J. Humphreys, R. Lan, S. Tao, Adv. Energy and Sustainability Res. 2021,2, 2000043. [8] A. Mittasch, Bemerkungen zur Katalyse, 1926. [9] B. Wilk, R. Pelka, W. Arabczyk, J. Phys. Chem. C 2017,121, 8548–8556. [10] S. Giddey, S. Badwal, A. Kulkarni, Int. J. Hydrog. Energy 2013,38, 14576–14594. [11] T. Kandemir, M. E. Schuster, A. Senyshyn, M. Behrens, R. Schloegl, Angew. Chem. Int. Ed. 2013,52, 12723–12726. [12] C. Smith, A. K. Hill, L. Torrente-Murciano, Energy Environ. Sci 2020,13, 331–344. [13] N. Lehnert, B. W. Musselman, L. C. Seefeldt, Chem Soc Rev 2021,50, 3640–3646. [14] B. S. Shridhar, Int. J. Microbiol. Res. 2012,3, 46–52. [15] J. P. Bellenger, R. Darnajoux, X. Zhang, A. M. Kraepiel, Biogeochemistry 2020,149, 53–73. [16] K. E. Luxem, A. M. Kraepiel, L. Zhang, J. R. Waldbauer, X. Zhang, Environ. Microbiol. 2020,22, 1397–1408. [17] R. Darnajoux, R. Bradley, J. P. Bellenger, Environ. Sci. Technol. 2022,56, 2760–2769. [18] P. C. D. Santos, R. Y. Igarashi, H. I. Lee, B. M. Hoffman, L. C. Seefeldt, D. R. Dean, Acc. Chem. Res. 2005,38, 208–214. [19] Y. Tanabe, Y. Nishibayashi, Chem. Soc. Rev. 2021,50, 5201–5242. [20] S. J. Forrest, B. Schluschaß, E. Y. Yuzik-Klimova, S. Schneider, Chem. Rev. 2021, 121, 6522–6587. [21] S. Hinrichsen, H. Broda, C. Gradert, L. Söncksen, F. Tuczek, Annu. rep. prog. chem. Sect. A: Inorg. Chem. 2012,108, 17–47. 107 Chapter 8 References [22] A. J. Martín, T. Shinagawa, J. Pérez-Ramírez, Chem 2019,5, 263–283. [23] F. Haber, The synthesis of ammonia from its elements, 1920. [24] C. Bosch, The development of the chemical high pressure method during the establishment of the new ammonia industry, 1932. [25] G. Soloveichik, Nat. Catal. 2019,2, 377–380. [26] K. H. R. Rouwenhorst, A. S. Travis, L. Lefferts, Sustain. Chem. 2022,3, 149–171. [27] G. J. Leigh in Vol. 1,2004, pp. 33–54. [28] B. Steininger, Anthr. Rev. 2021,8, 262–279. [29] T. Kandemir, M. E. Schuster, A. Senyshyn, M. Behrens, R. Schlögl, Angew. Chem. Int. Ed. 2013,52, 12723–12726. [30] A. Holleman, N. Wiberg, Anorganische Chemie,Vol. 103,2017, pp. 751–754. [31] G. Marnellos, M. Stoukides, Science 1998,282, 98–100. [32] G. Ertl, Angew. Chem. Int. Ed. 2008,47, 3524–3535. [33] M. Appl in John Wiley and Sons, Ltd, 1999, pp. 9–63. [34] P. H. Emmett in (Eds.: R. I. D. Edmund, Jaffee), Springer US, 1975, pp. 3–34. [35] G. Ertl, Adv. Catal. 2000,45, 1–69. [36] G. Ertl, Catal. Rev. Sci. Eng. 1980,21, 201–223. [37] H.-P. Jia, E. A. Quadrelli, Chem. Soc. Rev. 2014,43, 547–564. [38] G. Ertl, J. Vac. Sci. Technol. 1983,1, 1247–1253. [39] G. Ertl, Angew. Chem. 1990,102, 1258–1266. [40] I. I. Cheema, U. Krewer, RSC Advances 2018,8, 34926–34936. [41] R. Shi, X. Zhang, G. I. Waterhouse, Y. Zhao, T. Zhang, Adv. Energy Mater. 2020,10. [42] A. Vojvodic, A. J. Medford, F. Studt, F. Abild-Pedersen, T. S. Khan, T. Bligaard, J. K. Nørskov, Chem. Phys. Lett. 2014,598, 108–112. [43] V. Kyriakou, I. Garagounis, A. Vourros, E. Vasileiou, M. Stoukides, Joule 2020,4, 142–158. [44] S. Giddey, S. P. Badwal, A. Kulkarni, Int. J. Hydrog. Energy 2013,38, 14576–14594. [45] D. Fowler, M. Coyle, U. Skiba, M. A. Sutton, J. N. Cape, S. Reis, L. J. Sheppard, A. Jenkins, B. Grizzetti, J. N. Galloway, P. Vitousek, A. Leach, A. F. Bouwman, K. Butterbach-Bahl, F. Dentener, D. Stevenson, M. Amann, M. Voss, Philos. Trans. R. Soc. Lond. 2013,368. [46] L. Y. Stein, M. G. Klotz, Curr. Biol. 2016,26, R94–R98. [47] G. J. D. Kirk, H. J. Kronzucker, Annals of Botany 2005,96, 639–646. 108 Chapter 8 References [48] H. Daims, E. V. Lebedeva, P. Pjevac, P. Han, C. Herbold, M. Albertsen, N. Jehmlich, M. Palatinszky, J. Vierheilig, A. Bulaev, R. H. Kirkegaard, M. V. Bergen, T. Rattei, B. Bendinger, P. H. Nielsen, M. Wagner, Nature 2015,528, 504–509. [49] S. Lax, D. P. Smith, J. Hampton-Marcell, S. M. Owens, K. M. Handley, N. M. Scott, S. M. Gibbons, P. Larsen, B. D. Shogan, S. Weiss, J. L. Metcalf, L. K. Ursell, Y. Vázquez-Baeza, W. V. Treuren, N. A. Hasan, M. K. Gibson, R. Colwell, G. Dantas, R. Knight, J. A. Gilbert, Science 2014,345, 1048–1052. [50] M. Könneke, A. E. Bernhard, J. R. D. L. Torre, C. B. Walker, J. B. Waterbury, D. A. Stahl, Nature 2005,437, 543–546. [51] M. Palatinszky, C. Herbold, N. Jehmlich, M. Pogoda, P. Han, M. V. Bergen, I. Lagkouvardos, S. M. Karst, A. Galushko, H. Koch, D. Berry, H. Daims, M. Wagner, Nature 2015,524, 105–108. [52] N. von Wirén, S. Gazzarrini, A. Gojon, W. B. Frommer, Curr. Opin. Plant Biol. 2000, 3, 254–261. [53] E. J. Bedmar, E. F. Robles, M. J. Delgado, The complete denitrification pathway of the symbiotic, nitrogen-fixing bacterium Bradyrhizobium japonicum, 2004. [54] H. Cypionka, Grundlagen der Mikrobiologie,Vol. 4, Springer Berlin Heidelberg, 2010, pp. 197–201. [55] W. G. Zumft, Microbiol. Mol. Biol. Rev. 1997,61, 533–616. [56] J. J. G. Moura, P. Gonzalez, I. Moura, G. Fauque in Vol. 1,2007, pp. 241–264. [57] J. P. Zehr, D. G. Capone, Science 2020,368. [58] S. D. Threatt, D. C. Rees, FEBS Letters 2022,597, 45–58. [59] A. M. Marcarelli, R. W. Fulweiler, J. T. Scott, Limnol. Oceanogr. Lett 2022,7, 1–10. [60] B. Bergman, G. Sandh, S. Lin, J. Larsson, E. J. Carpenter, FEMS Microbiol. Rev. 2013,37, 286–302. [61] D. G. Capone, J. A. Burns, J. P. Montoya, A. Subramaniam, C. Mahaffey, T. Gunderson, A. F. Michaels, E. J. Carpenter, Global Biogeochem Cycles 2005,19, 1–17. [62] H. W. Paerl, Phycologia 1996,35, 25–35. [63] J. D. Noar, J. M. Bruno-Bárcena, Microbiology 2018,164, 421–436. [64] W. A. Bulen, J. R. Lecomte, PNAS 1966,56, 979–986. [65] Y. Hu, M. W. Ribbe, J. Biol. Inorg. Chem. 2014,19, 731–736. [66] H. Paerl, F1000Research 2017,6. [67] G. W. Niven, N. W. Kerby, P. Rowell, R. H. Reed, W. D. Stewart, J. Phycol. 1987, 22, 411–416. 109 Chapter 8 References [207] F. Meng, S. Kuriyama, H. Tanaka, A. Egi, K. Yoshizawa, Y. Nishibayashi, Angew. Chem. Int. Ed. 2021,60, 13906–13912. [208] S. Kuriyama, S. Wei, T. Kato, Y. Nishibayashi, Molecules 2022,27. [209] J. Krahmer, H. Broda, C. Näther, G. Peters, W. Thimm, F. Tuczek, Eur. J. Inorg. Chem. 2011, 4377–4386. [210] L. Söncksen, C. Gradert, J. Krahmer, C. Näther, F. Tuczek, Inorg. Chem. 2013,52, 6576–6589. [211] L. Söncksen, Dissertaion, 2013. [212] T. Weyrich, J. Krahmer, T. A. Engesser, C. Näther, F. Tuczek, Dalton Trans. 2019, 48, 6019–6025. [213] H. Broda, S. Hinrichsen, J. Krahmer, C. Näther, F. Tuczek, Dalton Trans. 2014,43, 2007–2012. [214] T. Weyrich, Dissertation, 2018. [215] H. Broda, Dissertation, 2013. [216] M. Pfeil, T. A. Engesser, J. Krahmer, C. Näther, F. Tuczek, Z. anorg. allg. Chem. 2021,647, 1778–1788. [217] M. Pfeil, T. A. Engesser, A. Koch, J. Junge, J. Krahmer, C. Näther, F. Tuczek, Eur. J. Inorg. Chem. 2020,2020, 1437–1448. [218] M. Pfeil, Dissertation, 2020. [219] J. Krahmer, Dissertation, 2011. [220] K. Klatt, Dissertation, 2008. [221] A.-C. Schnoor, Dissertation, 2015. [222] S. Dommaschk, Dissertation, 2016. [223] K. Grund, Unpublished Results, 2017. [224] K. Klatt, G. Stephan, G. Peters, F. Tuczek, Inorg. Chem. 2008,47, 6541–6550. [225] T. A. George, M. A. Jackson, B. B. Kaul, Polyhedron 1991,10, 467–470. [226] T. A. George, M. A. Jackson, Inorg. Chem. 1988,27, 924–926. [227] S. Hinrichsen, A. Kindjajev, S. Adomeit, J. Krahmer, C. Näther, F. Tuczek, Inorg. Chem. 2016,55, 8712–8722. [228] A. Kinjajev, Unpublished Results, 2017. [229] N. Stucke, J. Krahmer, C. Näther, F. Tuczek, Eur. J. Inorg. Chem. 2018,2018, 5108–5116. [230] J. Junge, T. A. Engesser, J. Krahmer, C. Näther, F. Tuczek, Z. anorg. allg. Chem. 2021,647, 822–831. 116 Chapter 8 References [231] T. A. Engesser, A. Kindjajev, J. Junge, J. Krahmer, F. Tuczek, Chem. Eur. J. 2020, 26, 14807–14812. [232] L. Alig, M. Fritz, S. Schneider, First-Row Transition Metal (De)Hydrogenation Catalysis Based on Functional Pincer Ligands, 2019. [233] S. Hinrichsen, A. C. Schnoor, K. Grund, B. Flöser, A. Schlimm, C. Näther, J. Krahmer, F. Tuczek, Dalton Trans. 2016,45, 14801–14813. [234] H. Zhang, R. Yuan, W. Wu, Y. Mo, Chem. Eur. J. 2020,26, 2619–2625. [235] A. G. Orpen, N. G. Connelly, Organometallics 1990,9, 1206–1210. [236] R. H. Crabtree, The Organometallic Chemistry of the Transition Metals,Vol. 6. Edition, John Wiley and Sons, 2013, pp. 19–36. [237] C. Gradert, J. Krahmer, F. D. Sönnichsen, C. Näther, F. Tuczek, Eur. J. Inorg. Chem. 2013, 3943–3955. [238] P. Dawson, Phyr. Chem.SoUd 1975,36, 1401–1402. [239] S. Bruderer, A. O. Benz, P. Stäuber, A. S. Davydov, Sov. Phys. Usp 1964,7, 393–448. [240] P. Ghosh, Solid State Commun. 1976,19, 639–642. [241] T. A. Gavrilko, G. O. Puchkovska, V. I. Styopkin, T. V. Bezrodna, J. Baran, M. Drozd, Ukr. J. Phys. 2013,58, 637–646. [242] K. Wang, D. Smith, Y. Zheng, Carbon Resour. Convers 2018,1, 2–31. [243] I. A. Amar, R. Lan, C. T. Petit, S. Tao, J Solid State Electrochem 2011,15, 1845–1860. [244] H. Miessner, J. Am. Chem. Soc 1994,116, 11522–11530. [245] H. Miessner, K. Richter, J Mol Catal A Chem 1999,146, 107–115. [246] D. L. Dubois, Inorg. Chem 1984,23, 2047–2052. [247] S. Ossinger, H. Naggert, L. Kipgen, T. Jasper-Toennies, A. Rai, J. Rudnik, F. Nickel, L. M. Arruda, M. Bernien, W. Kuch, R. Berndt, F. Tuczek, J. Phys. Chem. C 2017, 121, 1210–1219. [248] T. Knaak, C. González, Y. J. Dappe, G. D. Harzmann, T. Brandl, M. Mayor, R. Berndt, M. Gruber, J. Phys. Chem. C . 2019,123, 4178–4185. [249] T. G. Gopakumar, F. Matino, H. Naggert, A. Bannwarth, F. Tuczek, R. Berndt, Angew Chem. Int. Ed. 2012,51, 6262–6266. [250] E. Ludwig, H. Naggert, M. Kalläne, S. Rohlf, E. Kröger, A. Bannwarth, A. Quer, K. Rossnagel, L. Kipp, F. Tuczek, Angew. Chem. 2014,126, 3063–3067. [251] A. Ulman, Chem. Rev. 1996,96, 1533–1554. [252] A. Badia, R. B. Lennox, L. Reven, Acc. Chem. Res 2000,33, 475–481. 117 Chapter 8 References [253] M. Singh, N. Kaur, E. Comini, J. Mater. Chem. C 2020,8, 3938–3955. [254] J. Sánchez-Bodón, J. A. D. Olmo, J. M. Alonso, I. Moreno-Benítez, J. L. Vilas-Vilela, L. Pérez-Álvarez, Polymers 2022,14, 165. [255] S. P. Pujari, L. Scheres, A. T. Marcelis, H. Zuilhof, Angew. Chem. Int. Ed. 2014,53, 6322–6356. [256] S. A. Jadhav, Cent. Eur. J. Chem. 2011,9, 369–378. [257] E. Fast, A. Schlimm, I. Lautenschläger, K. U. Clausen, T. Strunskus, C. Spormann, T. K. Lindhorst, F. Tuczek, Chem. Eur. J. 2020,26, 485–501. [258] B. Baisch, D. Raffa, U. Jung, O. M. Magnussen, C. Nicolas, J. Lacour, J. Kubitschke, R. Herges, J. Am. Chem. Soc. 2009,131, 442–443. [259] R. Löw, T. Rusch, F. Röhricht, O. Magnussen, R. Herges, Beilstein J. Org. Chem. 2019,15, 1485–1490. [260] H. Jacob, S. Ulrich, U. Jung, S. Lemke, T. Rusch, C. Schütt, F. Petersen, T. Strunskus, O. Magnussen, R. Herges, F. Tuczek, Phys. Chem. Chem. Phys. 2014,16, 22643–22650. [261] A. Schlimm, N. Stucke, B. M. Flöser, T. Rusch, J. Krahmer, C. Näther, T. Strunskus, O. M. Magnussen, F. Tuczek, Chem. Eur. J. 2018,24, 10732–10744. [262] F. Petersen, I. Lautenschläger, A. Schlimm, B. M. Flöser, H. Jacob, R. Amirbeigiarab, T. R. Rusch, T. Strunskus, O. Magnussen, F. Tuczek, Dalton Trans. 2021,50, 1042– 1052. [263] R. Huisgen, 1,3-Dipolar cycloaddition chemistry. A. Padwa, Ed., Wiley, 1984, pp. 1– 176. [264] E. Hrishikesan, R. Manjunath, P. Kannan, J Solution Chem 2016,45, 907–919. [265] L. R. Hoover, T. Pryor, J. A. Weitgenant, P. E. Williams, B. N. Storhoff, J. C. Huffman, Phosphorus Sulfur Silicon Relat. Elem. 1997,122, 155–166. [266] K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Letters 1975, 4467–4470. [267] K. Leonard, M. Nelen, M. Raghu, M. R. Detty, J. Heterocycl. Chem. 1999,36, 707–717. [268] G. Wittig, W. Böll, Chem. Ber. 1962,95, 2526–2534. [269] A. N. Nesmeyanov, V. V. Krivykh, V. S. Kaganovich, M. L. Rybinskaya, J. Organomet. Chem. 1975,102, 185–193. [270] B. B. Nicholls, M. C. Whiting, J. Chem. Soc. 1959,69, 551–556. [271] C. D. Hoff, J. Organomet. Chem. 1985,282, 201–214. [272] F. A. Cotton, J. E. McCleverty, J. E. White, R. B. King, A. F. Fronzaglia, M. A. Bisnette in Academic Press, 1967, pp. 45–47. [273] K. U. Clausen, Current Works, 2023. 118 Chapter 8 References [274] S. G. Keskin, J. M. Stanley, A. H. Cowley, Polyhedron 2017,138, 206–217. [275] C. Vericat, M. E. Vela, G. Benitez, P. Carro, R. C. Salvarezza, Chem. Soc. Rev. 2010, 39, 1805–1834. [276] J. W. Ciszek, M. P. Stewart, J. M. Tour, J. Am. Chem. Soc. 2004,126, 13172–13173. [277] J. W. Ciszek, M. P. Stewart, J. M. Tour, J. Am. Chem. Soc. 2004,126, 13172–13173. [278] K. Bourumeau, A. - C. Gaumont, J. - M. Denis, J. Organomet. Chem. 1997,529, 205–213. [279] K. Zhu, P. D. Achord, X. Zhang, K. Krogh-Jespersen, A. S. Goldman, J. Am. Chem. Soc. 2004,126, 13044–13053. [280] P. Stössel, H. A. Mayer, C. Maichle-Mössmer, R. Fawzi, M. Steimann, Inorg. Chem. 1996,35, 5860–5867. [281] M. Ohff, J. Holz, M. Quirmbach, A. Börner, Synthesis 1998,10, 1391–1415. [282] B. Rapp, J. E. Drake, Inorg. Chem. 1973,12, 2868. [283] N. A. Bondarenko, K. V. Tcarkova, S. K. Belus’, O. I. Artyushin, Phosphorus Sulfur Silicon Relat. Elem. 2021,196, 902–910. [284] N. Michaelis, Masterthesis, 2021. [285] T. Leischner, A. Spannenberg, K. Junge, M. Beller, Organometallics 2018,37, 4402– 4408. [286] N. F. Both, A. Spannenberg, K. Junge, M. Beller, Organometallics 2022,41, 1797– 1805. 119 Chapter 9 Appendix 1 Supporting Information Tungsten and Molybdenum Dinitrogen Complex Supported by a Pentadentate Tetrapodal Phosphine Ligand: Comparative Spectroscopic, Electrochemical and Reactivity Studies Jannik Junge†a, Sven Froitzheim†a, Tobias A. Engessera, Jan Krahmera, Christian Näthera, Nicolas Le Poul*b, Felix Tuczek*a Content page [WCl3(κ3-P2MePP2Ph] 2 [W(N2)(P2MePP2Ph)] (2) 4 Electrochemistry of [Mo(N2)(P2MePP2Ph)] (1) and [W(N2)(P2MePP2Ph)] (2) 12 1H-DOSY NMR spectrum of [W(N2)(P2MePP2Ph)] (2) 13 [W(NNH2)(P2MePP2Ph)](BArF)2 (3-BArF) 14 [W(NNH2)(P2MePP2Ph)][Al(pftb)4]2 (3-Al(pftb)4) 24 Catalytic experiments 24 Cartesian coordinates of [W(N2)(P2MePP2Ph)] (2) and [W(NNH2)(P2MePP2Ph)]2+ (3) 25 a. Institut für Anorganische Chemie, CAU Kiel, Max-Eyth-Str. 2, 24118 Kiel. Email: [email protected] b. Laboratoire de Chimie, Électrochimie Moléculaires et Chimie Analytique (UMR CNRS 6521) Université de Bretagne Occidentale, 6 Avenue Le Gorgeu, 29238 Brest, France Email: [email protected] Electronic Supplementary Material (ESI) for Dalton Transactions. This journal is © The Royal Society of Chemistry 2022 121 Chapter 9 Appendix 2 [WCl3(κ3-P2MePP2Ph] Figure S1. X-band (9.8 GHz) EPR spectra of thf solutions of [WCl3(κ3-P2MePP2Ph] at room temperature (left) and 77 K (right). The X-band EPR spectra were recorded on a Bruker EMX Plus spectrometer with dual mode cavity, 2.0 mW microwave power and modulation amplitude of 1 mT. Figure S2. IR and Raman spectrum of [WCl3(κ3-P2MePP2Ph]. 122 3 Table S1. Experimental frequencies of the Raman and IR bands of [WCl3(κ3-P2MePP2Ph] [cm-1]. [WCl3(κ3-P2MePP2Ph] (2) Raman IR Assignment Raman IR Assignment - 3067 (w) - - 945 (m) CC 3054 (m) 3051 (w) ν(CH) - 917 (m) CC 2982 (w) - νas(CH3/CH2) - 880 (m) - - 2954 (sh) - - 843 (w) - 2913 (s) 2919 (m) νs(CH3/CH2) - 805 (vw) - - 2867 (sh) - 786 (w) - - - 2801 (w) - - 739 (s) - 1586 (s) 1589 (m) CC 688 (w) 693 (vs) - 1573 (w) 1568 (w) - - 639 (vw) - - 1551 (vw) - 619 (w) 617 (w) - - 1480 (m) - - 577 (vw) - 1455 (vw) 1455 (vw) - - 543 (w) - - 1430 (m) PC - 508 (s) - 1414 (w) 1412 (w) PC - 479 (m) - - 1381 (w) - - 442 (vw) - - 1330 (vw) - - 425 (w) - 1306 (vw) 1297 (w) - 400 (vw) 404 (vw) - - 1279 (w) - - 362 (vw) - - 1260 (vw) - - 340 (sh) - - 1241 (w) - 328 (w) 323 (sh) - 1210 (vw) - - - 294 (s) - 1186 (w) 1182 (w) - - 271 (vs) - 1160 (w) 1156 (w) - 258 (w) 252 (sh) - - 1119 (w) - - 221 (vw) - 1100 (w) 1096 (m) - - 208 (vw) - 1073 (vw) 1069 (w) - - 178 (w) - 1028 (m) 1025 (w) - 150 (vw) 158 (w) - 1000 (vs) 1000 (w) CC - 133 (m) - 960 (w) - - - 120 (w) - vw = very weak, w = weak, m = medium, s = strong, vs = very strong; ν = valence, δ = deformation 123 Chapter 9 Appendix 4 [W(N2)(P2MePP2Ph)] (2) NMR spectroscopy Figure S3. 1H NMR spectrum of 2 in benzene-d6. Figure S4. Enlargement of the aliphatic region of the 1H NMR spectrum of 2 in benzene-d6 and assignment of the signals. 124 5 Figure S5. Enlargement of the aromatic region of the 1H NMR spectrum of 2 in benzene-d6. Figure S6. 13C NMR spectrum of 2 in benzene-d6. 125 Chapter 9 Appendix 12 Supplementary electrochemical and spectroelectrochemical data for [W(N2)(P2MePP2Ph)] (2) and [Mo(N2)(P2MePP2Ph)] (1). Figure S13. A) Plot of ipa(1) vs. v1/2 from CV data obtained for [W(NNH2)(P2MePP2Ph)] (0.4 mM) in THF/NaBPh4 20 mM, red curve: linear fit; B) CV of [Mo(N2)(PMe2PPPh2)] (1 mM) at Pt working electrode (diam. 1 mm) for v = 05 V s-1 in THF/NaBPh4 20 mM.; C) Infrared spectra of [W(N2)(PMe2PPPh2)] (15 mM) in THF/NaBPh4 20 mM recorded during in-situ spectroelectrochemical measurements before (black) and after oxidation at Epa(1) (red), then returning back to the initial potential (blue); (D) The same as (C) except that oxidation at Epa(1) (red) is followed by oxidation at Epa(2) (green), then back-reduction to Epc(1) < E < Epa(2) (blue) and finally Epc(1) (cyan). 132 13 1H-DOSY NMR experiment of [W(N2)(P2MePP2Ph)] (2) The 1H-DOSY NMR was measured with 13 mg (14.6 µmol) of [W(N2)(P2MePP2Ph)] (2) and 17 mg (49.7 µmol) NaBPh4 in 0.5 ml deuterated THF. The resulting value for the diffusion coefficient was 9.5∙10-6 cm2 s-1. Figure S14. 1H-DOSY NMR spectrum of [W(N2)(P2MePP2Ph)] (2). 133 Chapter 9 Appendix 14 Vibrational spectroscopy of [W(NNH2)(P2MePP2Ph)](BArF)2 (3-BArF) Figure S15. a) IR and b) Raman spectra of (3-BArF) (black) and 15N2-3-BArF (red). The enlargement shows the N-H vibrational frequency and its shift when labelling with 15N isotope. 134 15 Table S7. Experimental frequencies of the Raman and IR bands of [W(NNH2)(P2MePP2Ph)](BArF)2 (3-BArF) [cm-1]. [W(NNH2)(P2MePP2Ph)](BArF)2 / [W(15N15NH2)(P2MePP2Ph)](BArF)2 NaBArF Assignment Raman IR Raman IR - 3312 (w) / 3308 (w) - - N-H 3074 (w) 3076 (vw) 3084 (m) - ν(CH) 3023 (vw) - 3029 (w) - νas(CH3/CH2) 2960 (w) 2960 (w) - 2967 (vw) - 2932 (m) 2924 (m) - - νs(CH3/CH2) 2906 (m) - - - - 2882 (sh) - 2885 (w) - - 2852 (w) 2853 (w) - - - - - 2637 (vw) 2641 (vw) - - - 2592 (vw) - - - - 2541 (vw) - - - 1790 (vw) - - - - - - 1629 (w) - 1611 (m) 1611 (m) 1611 (m) 1612 (m) CC 1593 (s) - 1596 (s) - CC 1523 (vw) - - - - 1464 (w) - 1467 (vw) - - - 1443 (w) - - - - - 1422 (vw) - - 1364 (s) 1352 (s) 1367 (s) 1355 (s) CF - - 1323 (vw) - - - 1272 (vs) 1272 (w) 1277 (vs) CC, CF - - 1197 (w) 1190 (sh) - - 1159 (m) 1161 (w) 1170 (m) CC, CF - - - 1139 (sh) - 1107 (w) 1114 (vs) 1113 (w) 1115 (vs) CC, CF - 1092 (sh) - - - - - - 1063 (s) - 1030 (w) - 1043 (w) - - 1003 (vs) 998 (w) 1000 (vs) 1000 (w) BC 135 Chapter 9 Appendix 16 - - - 964 (w) - 952 (w) - - - - 939 (vw) 934 (w) 935 (w) 945 (m) - - - - 933 (m) - 914 (vw) 918 (vw) - - - - 886 (s) 892 (w) 886 (vs) CC 840 (vw) 838 (s) 837 (vw) 838 (s) CC - 809 (m) - - - 801 (s) - 801 (s) - CC 745 (w) 743 (m) 744 (m) 742 (m) - 704 (s) 711 (s) 702 (s) 708 (vs) CC - 698 (w) - - - 689 (w) 680 (s) - 678 (vs) CC 675 (w) 669 (s) 673 (m) 670 (vs) - 617 (vw) 617 (vw) - - - - 609 (vw) - 609 (vw) - - 580 (w) 583 (w) 582 (w) - - 568 (vw) / 554 (w) - - W-N - 516 (w) - - - - 507 (w) - 505 (vw) - - 488 (w) - - - - 449 (m) - 449 (m) - - 403 (w) 410 (m) 401 (w) - - 392 (w) 389 (w) 386 (w) - - 380 (vw) - - - - 366 (m) - 366 (m) CC, CF - 353 (sh) 354 (w) - - - 331 (vw) - - - - - 321 (w) 317 (vw) - 287 (w) 283 (w) 293 (m) 287 (w) CC 262 (w) 259 (w) 261(w) 259 (w) - 244 (vw) 246 (vw) - - - 235 (w) - 238 (s) - - - - 216 (w) - - vw = very weak, w = weak, m = medium, s = strong, vs = very strong; ν = valence, δ = deformation 136 17 [W(NNH2)(P2MePP2Ph)](BArF)2 (3-BArF) NMR spectroscopy Figure S16. 1H NMR spectrum of 3-BArF in diethylether-d10. Figure S17. Enlargement of the aliphatic region of the 1H NMR spectrum of 3-BArF in diethylether-d10 and assignment of the signals. 137 Chapter 9 Appendix 18 Figure S18. Enlargement of the aromatic region of the 1H NMR spectrum of 3-BArF in diethylether-d10 and assignment of the signals Figure S19. 13C NMR spectrum of 3-BArF in diethylether-d10. 138 19 Figure S20. Enlargement of the aliphatic region of the 13C NMR spectrum of 3-BArF in diethylether-d10 and assignment of the signals. Figure S21. Enlargement of the aromatic region of the 13C NMR spectrum of 3-BArF in diethylether-d10 and assignment of the BArF-signals. 139 Chapter 9 Appendix 20 Figure S22. Enlargement of the aromatic region of the 13C NMR spectrum of 3-BArF in diethylether-d10 and assignment of the diphenylphosphine signals. 140 21 Figure S23. 31P-31P-COSY spectrum of 3BArF in diethylether-d10. Figure S24. 1H-31P-HMBC spectrum of 3-BArF in diethylether-d10. 141