1
MSc Chemistry
Molecular Sciences
Literature Thesis
Photochemically activated Sulphur chelated Grubbs Olefin
Metathesis Catalysts
by
Kim Zijderveld
12369160
June 2020
12 EC
March 15
th, 2020 – June 18
th, 2020
Supervisor and Examiner:
2
ndExaminer:
Prof. dr. B. de Bruin
prof. dr. C.J. Elsevier
Van 't Hoff Institute for Molecular Sciences
2
Abstract
A relatively new method to activated sulphur chelated Grubbs olefin metathesis catalysts is photochemical activation. This study set out to determine if photochemical activation of these catalysts is superior to other forms of activation. This is done by comparing the activation method, the scope and the potential of the reaction. The most significant finding is a larger scope and higher conversion of the thermal activation method. Based on this result, the thermally activated catalysts are best suited for most olefin metathesis reactions. An exception is the potential the photochemically activated catalysts offer to the metathesis polymerisation reaction, as the reaction could be activated and deactivated by light, creating an external on/off switch. This could be beneficial for the polydispersity index, which expresses the size distribution. The lowering of the polydispersity with the on/off switch is possible because olefin metathesis polymerisation a living polymerisation reaction. Another benefit of the photochemically activated catalyst is the potential towards activation by sunlight. This would be beneficial to the green aspects of the olefin metathesis. Thus, the findings of this study suggest the photochemically activated catalysts are better suited for specific applications of the polymerisation reaction and the green aspect of the metathesis reaction.
3
Content
Abstract ... 2
Introduction ... 4
The olefin Metathesis reaction ... 5
Introduction ... 5
Mechanism ... 5
Ring closing metathesis ... 6
Ring opening metathesis ... 6
Ring opening metathesis polymerisation... 8
The olefin Metathesis catalysts ... 9
Initiation step ... 9
Propagation ... 13
Aim ... 15
Photochemically activated Grubbs catalysts ... 16
Isomerisation ... 16
Mechanism of the activation ... 18
Reversion of the trans configuration ... 19
Deactivation ... 19 Activation by heat ... 20 Reaction scope ... 22 RCM ... 22 ROMP ... 24 Selectivity ... 25 Potential ... 26 Polymerisation ... 26 Sustainability ... 26 Conclusion ... 28 References ... 29
4
Introduction
Light has become a great potential energy source explored by chemists in the field of photochemistry and the study of natural processes1,2,3. A recent new field of exploring the utility of light is in
combination with the olefin metathesis reaction. In 2005 Yves Chauvin, Richard Schrock, and Robert Grubbs won the Nobel prize for development of the olefin metathesis reaction4. Since then the
metathesis reaction has been even further researched. The combination of this useful and widely applicable reaction with light has among other things produced a photochemically activated Grubbs sulphur chelated catalyst5,6,7. The potential application of photochemically activated catalyst are
large. They can be utilised in paint, so the paint cures by exposure to light, but not yet during storage. As the exposure to light only occurs during painting and not in the container it will be distributed in. But also in block polymers, to activate the production of one block, as second block in a one pot reaction. Since the photochemically activated reaction can be turned on and off, the controllability of the blocks in the polymers would be external. To determine the full potential of this reaction more research on photochemically activated metathesis catalyst is necessary.
Besides the photochemical application of the olefin metathesis reaction, the reaction itself is well established. Throughout the years the research on the reactions and their applications have become more diffused into all the fields of chemistry. Recently, the majority of the catalysts used in the industry are Grubbs catalysts which are a ruthenium carbene metathesis catalyst discovered by Grubbs in 19928,9. For example, the syntheses of BILN-206110, simeprevir11 and vaniprevir12 are done
by implementation of Grubbs olefin metathesis reactions. The Grubbs type catalyst has been proven to be photochemically activatable6,7. However, it is not yet clear whether this activation method has
advantages which can be practically applied over other types of activation. This literature thesis discusses the usability of photochemically activated complexes compared to a more traditional method as thermal activation, to conclude if there are any benefits of using photochemical activation over using thermal activation.
5
The olefin Metathesis reaction
Introduction
The olefin metathesis reaction is a versatile reaction for the formation of carbon-carbon double bonds by cleaving carbon-carbon bonds and interchanging the carbene fragment of different alkenes
13,14. As the reaction has a broad range and is often utilised, this reaction has multiple sub-categories.
Some of these will be discussed below as well as an account of the mechanism.
Mechanism
The mechanism for the olefin metathesis reaction is relatively simple. The reaction starts with a metal carbene complex, followed by the addition of an alkene and the creation of a metallacycle intermediate, which eliminates a new alkene via a cycloreversion15,16. As a consequence, the carbene
on the metal changes too. The mechanism as shown in figure 1, and is referred to as the Chauvin mechanism. The Chauvin mechanism is well established and utilises a metal carbene also named metal alkylidene. An incoming alkene coordinates to metal (B), followed by the metallacycle intermediate (C). By opening up the metallacycle intermediate a new carbene and alkene are formed (D).
Figure 1. The reaction mechanism for olefin metathesis cross-coupling reaction.
In the Chauvin mechanism, it is proposed that the d-orbital of the transition metal is decreasing the energy required to perform this reaction, shown in figure 217. This is supposed to be caused by the
metal donating electrons in the antibonding orbital, thus weakening the alkene bond and lowering the activation energy of the alkene. The above described reaction shown in figure 1 is a cross-coupling olefin metathesis reaction in its simplest form, where the substituent on the alkenes are crossed over. Besides the cross-coupling olefin metathesis reaction there are other classes of olefin metathesis reactions, namely ring closing metathesis, ring opening metathesis and the ring opening metathesis polymerisation.
Figure 2. Metal double bond orbital interaction when the olefin is coordinated to the transition metal 2.
C B A = A D
6
Ring closing metathesis
The ring closing metathesis reaction (RCM) is an intramolecular reaction between two double bonds forming a ring, which is most successful with 5, 6, and 7 membered rings18. Normally, the rate of the
ring formation follows the trend of 5>6>3>7>4>8-1019. However, this is based on the assumption
that the 3 and 4 membered rings are able to be formed, due to the low entropy. Therefore, the chance of the two alkenes meeting is increased. The low entropy effect does not apply for the metathesis reaction because in this case the alkenes are forced in a pre-conformation. The increase of strain be the formation of the 3 and 4 membered rings combined with the lack of stabilising effects from the entropy, makes the formation of the 3 and 4 membered rings doubtful. Taking entropy and strain into account the 5-10 membered rings are just as likely to form as in organic reactions. Thus, the presumable reaction rate becomes 5-6>7>8-10 16. The stereochemistry of the ring closing
metathesis is predetermine. As the ring is required to be in a cis position in order to form, the side groups will rotate into a cis position too19.
An important factor for lowering of the energy barrier of the ring closing metathesis is the Thorpe-Ingold effect. This effect explains the increasing rate of cyclization of substituted substrates. The effect is due to the bond angle between the substituents, which becomes larger when the ring is formed. Since the ring forces the carbon-carbon bonds to be strained with a lower bond angel, there is more space for the substituents. This effect is more noticeable with large substituents20,21.
The Thope-Ingold effect can be considered to have a positive contribution to the ring closing metathesis. However, if the metal has a limited amount of space due to bulky ligands, the extra substituents on the substrate could have an adverse effect on the reaction rate22.
Figure 3. Schematic representation of the Thorpe-Ingold effect.
Ring opening metathesis
The ring opening metathesis (ROM) utilizes a ring with a double bond and a separate alkene to open the ring. This reaction is most effective with small rings from 3 to 4 carbons, bridging rings, and large rings from 8 to 11 carbons23,24. This is due to strain in small rings and the instability of large rings20,25.
As the rings themselves are relatively reactive compared to 5 or 6 membered rings, the olefin metathesis reaction is more efficient.
7
Figure 4. Explanation of the stereoselectivity of the ring opening metathesis reaction13.
The stereochemistry of the ring opening metathesis is different from the ring closing metathesis. For the ring opening metathesis the configuration of the product is mostly the configuration of the reactant, with the exception of a R group present on the metal carbene in which case the configuration is trans13. The determination of the configuration is explained within the mechanism of
the olefin metathesis as shown in figure 413,26. An example of this predetermined configuration is
found with highly substituted cyclopentanes and furan derivatives as reactants. The stereoselectivities of the highly substituted cyclopentanes and furan derivatives are low with E:Z ratios ranging from 5:1 to 1:618. One of the reactions is shown in figure 5, which correspond to the
stereoselectivity of the third reaction shown in figure 4.
Figure 5. Ring opening metathesis reaction with stereoselectivity.
8
Ring opening metathesis polymerisation
The ring opening metathesis polymerisation (ROMP) is a normal ring opening metathesis where the double bond newly created becomes part of a polymer backbone27. The metathesis polymerisation is
achieved via a living polymerisation mechanism28,18. In order to consider a polymerisation living, the
following aspects must be fulfilled: a full and fast initiation of the monomer, the degree of polymerization is excepted to be linear to the starting concentration of the monomer, and the polydispersity index (PDI) of the polymer should be smaller than 1.529. In essence, living
polymerisation means that the polymer is growing onto the metal. Within this mechanism the first ring will coordinate onto the metal. Presumably, followed by opining this coordinated ring before a new ring can coordinated. The carbene of the opened ring will probably help opening the second ring by cross-coupling, thereby forming the polymer. As proof of the living polymerisation a PDI of 1.1 has been reported, which the authors argue could be decreased even futher30. Thus, they concluded the
living polymerisation the correct categorisation for the metathesis reaction.
9
The olefin Metathesis catalysts
There are two types of olefin metathesis catalysts, the early transition metals and the Ru based catalysts31,32,33. The Ru based catalysts are more utilised in the industry than the early transition
metals 9. The Ru catalysts are more inert towards most common functional groups34. Besides the high
tolerance for active groups, the Ru catalysts also have high kinetic stability towards water and air, and are highly active35. Grubbs inspired multiple different complexes which feature as the central
metal Ruthenium. There are different generations of the Grubbs catalyst. The higher the generation of the catalyst, the more inert the catalysts becomes to air and water. This is achieved by changing the ligands shown in figure 6. Besides the first generation and second generation catalysts30,36,
several other catalysts have been produced. Examples of those catalysts are the Hoveyda-catalyst37
and the Hoveyda-Grubbs catalyst38. As there are a lot of olefin metathesis catalyststhere has been a
large shift towards development of well-defined and specific catalysts6. In this chapter the different
types of catalysts will be compared in the separate stages of the metathesis reaction. The metathesis reaction occurs via an initiation, propagation, and restoration of the catalyst. As the restoration of the catalyst is the reversal of the initiation step it is not separately discussed.
Figure 6. Structure A is an example of a first generation Grubbs catalyst, structure B is an example of a second generation Grubbs catalyst and structure C is an example of the Hoveyda-Grubbs catalyst. The X in structure C represents a chelated S,N
or O atom.
Initiation step
The Grubbs catalyst could be considered a pre-catalyst39. A pre-catalyst or initiator catalyst is a
catalyst which requires an external stimulus to become active, which can be achieved in multiple ways. For example with acid40, oxidation41, or the use of chloride42. The initiation step is essentially a
substitution reaction between one of the ligands and the double bond which is needed to perform an olefin metathesis reaction. The substitution of ligands can be done by an associative or dissociative mechanism14. With an associative mechanism a alkene will first coordinate creating an 18 electron
complex, before the dissociation a ligand upon excitation as shown in figure 7A. The dissociative mechanism would first have a ligand dissociation, before association of the alkene, which would provide a 14 electron complex intermediate as shown in figure 7B43. In 2001, Sanford, Jennifer, and
Grubbs43 established the dissociative substitution as the correct mechanism for the initiation step of
10
Figure 7.Reaction A is an associative initiation mechanism and reaction B is a dissociative initiation mechanism.
The initiation is believed to be the rate determining step for the overall olefin metathesis reaction, this step could be improved through the optimisation of the ligands44,45. The reason for the
optimisation of the ligands instead of changing the olefin concentration is the seemingly independence of the olefin concentration on the rate of the reaction45. However, the independence
of the olefin concentration is not the case under all circumstances. The dependency is relevant when the rate of olefin coordination is larger than the boomerang effect43. In the boomerang effect the
phosphine will dissociate and a phosphine will associate. Therefore, the 14 electron complex becomes a 16 electron complex again before the olefin is coordinated. So, an optimal ligand is weakly bound to the metal to lower the boomerang effect and thus, increasing the reaction rate.
The first generation catalyst has a phosphine as ligand as shown in figure 6 complex A8. In the first
generation the dissociation of phosphine ligand will be the limiting factor of the rate determining step. Interestingly, the PPh3 was less suitable as a ligand than the PCy3. The replacement led to 50
times faster phosphine exchanges43. The reason for the difference is unknown, but the PCy
3 is mostly
used in the first generation because it dissociates faster46,47. To increase the rate of the dissociation
step an enhancement of the trans effect could be used via alternative ligands43,45. The trans effect is
an electronic effect where ligands labialises ligands at the trans position by donation of electron density trough the metal.
11
The N-heterocyclic carbene (NHC) second-generation Grubbs catalysts, as shown in figure 6 complex B48 is currently more used than the first generation with its two phosphine groups49,50. There is not
much known about the rate of dissociation for the phosphine trans to the NHC compared to the first generation catalyst. However, the second-generation Grubbs catalyst is known to be more efficient in coordination of the olefin51. The effectivity of the olefin coordination is due to the affinity for olefins
over the free phosphine43, which decreases the boomerang effect. The reduction of the boomerang
effect is possibly due to the electronic properties, as NHC groups are known electron donors to trialkylphosphines though the metal52,53. As phosphorous ligands are capable of sigma donate and pi
back donate, where the pi back donation destabilises the bond, the phosphorous-metal bond will weaken11. The pi back donation will increase when the metal is more electron rich. Thus the NHC
ligand was an improvement on the activation step. The NHC ligands have an identical backbone but are alterable by the side groups. There appears to be a correlation between an increase in size of the ligands and an increase in the initiation rate54.
Besides the first and second generation catalysts, there is the Hoveyda-Grubbs catalyst shown in figure 6 complex C37,38. The Hoveyda-Grubbs catalyst does not have any phosphine ligands. Instead
the Hoveyda-Grubbs catalyst utilises oxygen, nitrogen, or sulphur to coordinate to the metal on position X as part of the chelated ligand in figure 655. For the Hoveyda-Grubbs catalyst the initiation
step is a little more complex, since the entire chelating ligand has to dissociated48. In contrast to the
12
Figure 10. DFT calculations for the initiation step48.
The dissociation of the chelated ligand takes place via a whole range of steps shown in figure 9. The overall dissociation process is assumed to be the rate determining step for the olefin metathesis reaction48. The DFT calculation shown in figure 10 reveals interesting aspects of this reaction the
most relevant two will be explained48. First, the dissociation of the chelated atom has the highest
energy barrier. This is comparable to the initiation mechanisms of the first and second generation catalyst, where the phosphine group is dissociating is rate determining. The high energy for the dissociation is overcome by a predicted rotation of the phenyl ring to create space for coordination of a olefin. Presumably, the NHC ligands used in the Hoveyda-Grubbs catalysts still favours the olefin over the phosphine as a ligand as discussed previously.
Secondly, the approach of the olefin has a surprising effect on the metal carbene. The phenyl ring of the chelated carbene seems to rotate perpendicular to its previous position. In this geometry, there is no pi-electron delocalisation possible between the carbene and the phenyl ring. The rotation in the perpendicular configuration to the chelated phenyl is necessary for formation of the ruthena(iv)cyclobutane due to the orbital orientation35. As shown in figure 11, the rotation of the
carbene is necessary for the possibility of orbital interaction of the two double bonds. When the carbene is configurated as on the left shown in figure 11, the orbitals are perpendicular to each other and the ruthena(iv)cyclobutane cannot be formed.
Figure 11. Orbital overlap for the rotated carbene and the normal coordinated carbene.
13
Solans-Monfort, Pleixats, and Sodupe48 stated that the main factor for the high energy of the total
chelated ligand dissociation energy barrier is due to the high barrier in the rotative step. This rotation is not unique to the chelated carbene of the Hoveyda-Grubbs catalyst but the impact is much larger due to the disruption of the delocalisation of the electrons. Since these authors48 concluded that the
rotation was the main factor for the high energy barrier, they neglect the dissociation of the chelated oxygen, nitrogen or sulphur. As the DFT calculation show the highest energy barrier for the dissociation of the chelated oxygen, nitrogen or sulphur, the rate will increase by a decrease in bond strength of the oxygen, nitrogen or sulphur ruthenium bond. Furthermore, the combination of the dissociation and the rotation of the carbene contribute majorly in the overall high energy barrier of the chelated carbene dissociation step, as shown in figure 10.
The dissociation of the chelated oxygen, nitrogen or sulphur ruthenium bond appears to be the highest barrier in the olefin metathesis process and its properties have been studied55. One of the
factors of a strong bond is the bond length as a longer bond is generally broken more easily. The trend in bond length for the Hoveyda-Grubbs catalyst is S>O>N55. This is assuming the atomic radii
do not affect the bond strength. Besides bond length of ruthenium and the chelated oxygen, nitrogen or sulphur, the ruthenium chelated carbene bond length changes by selecting an oxygen, nitrogen or sulphur as chelated atom. As the chelated carbene needs to dissociate to create the active olefin metathesis catalyst, the assumption is made that increasing ruthenium-chelated bond length increases the initiation rate. The bond length of the carbene follows the same trend S>O>N55, making
the sulphur chelated catalysts the fastest in the dissociation step.
Propagation
The propagation step has relatively low energy barriers compared to the initiation step of the metathesis reaction48. However, the overall rate of the reaction is not directly proportional to the
rate of the activation step43. Although the activation step is the rate-determining step, some catalysts
which perform best in the initiation step will perform worst in the propagation step. The second generation Grubbs catalyst and Hoveyda-Grubbs catalyst are lower in energy than the first generation Grubbs catalyst presumably because the NHC stabilises the electron rich ruthenium43. The
lower the energy the more stable a complex becomes which might be unfavourable for the propagation step. Thus, it can be assumed that the first generation Grubbs catalyst might preform more efficient compared to the second generation Grubbs catalyst and the Hoveyda-Grubbs catalyst. However, in the over all reaction this effect will most likely be undermined by the preference of the second generation Grubbs catalyst and the Hoveyda-Grubbs catalyst in the initiation step. In addition to changing the catalyst, a change of substrate will impact the reaction too. Unsurprisingly, the propagation step is faster with smaller substrates, which will enhance the overall speed of the reaction57.
14
15
Aim
The aim of this paper will be to conclude whether the relatively new photochemical activation can overthrow the established thermal activation of the sulphur chelated Hoveyda-Grubbs catalysts based on the practical application. To determine the practical application an estimation on the efficiency, scope, and potential of these catalysts will be established.
16
Photochemically activated Grubbs catalysts
One of the most recent developments made is the activation of a complex via light58,7.
Photochemically activated catalysis can have bread interpretation of the definition. One of these interpretations include the redox catalytic cycle. In a redox catalytic cycle the light is used to excite an electron, making the excited electron more reactive towards redox reactions. As the excitation is repeated per cycle59,the photochemical activation of this reaction is debatable. The reaction could be
categorised as light stimulated instead of photochemically activated. As this paper is focusing on photochemically activated catalysts the redox reactions will not be included.
A photochemically activated catalyst is defined in this paper as a catalyst which is once activated from being a pre-catalyst to an activate catalyst. The activation of the pre-catalyst can be achieved by a light induced isomerisation. In the photochemically activated Grubbs catalysts section the context will be confined to the sulphur-chelated isomerisation mechanism, the deactivation of these complexes, and steric effects on the rate.
Isomerisation
A catalyst which is acknowledged to be a photochemically activated catalyst, is the photo isomerization of a sulphur chelated ruthenium benzylidene complex as shown in figure 137. The trans
configuration of the complex has been shown to be more active than the cis configuration of the complex21,60. As an example, for the RCM of multiple substrates the trans isomer complex was
reported to have a moderate activity at room temperature, where the cis isomeric complex was active only at high temperature61. The little activity observed for the cis complex at high temperature
was attributed to a small isomerisation towards the trans complex, which would make the trans complex the only active complex towards the olefin metathesis reaction. This is supported by other sources claiming that the cis complex is inactive21, some even describe the cis isomer as a
pre-catalyst58.
Figure 13. Top: Ru complex and the activation of the Ru catalyst7. Bottom: DFT calculated isomers and isomer barrier21.
17
The energy barriers of isomerisation of chelated complexes are typically high39, as the free rotation is
hindered62. The free rotation is hindered because the chelated ligand is attached in two places on the
metal. The high activation energy for the chelated complexes prevents them from isomerisation at ambient conditions. However, the isomerisation can be achieved by the use of light as a stimulus, which is proven to be successfully accomplished21.
The cis-configuration is supposed to be the thermodynamic product whereas the trans-configuration is said to be the kinetic product21. However, DFT calculation show that the energy barrier from the 14
electron complex towards the trans-complex is a little higher than the energy barrier towards the cis complex shown in figure 1321. This would make this the cis-configuration not only the thermodynamic
product but also the kinetic product. Since the trans-configuration exists after irradiation58, it is
plausible that the exact amount of energy for the trans barrier is provided by the light. Thus, making it possible to isomerise the complex.
The isomerisation as activation strategy appears to be effective for different side groups on the sulphur atom58. Even for large side groups on the sulphur atom it will proceed via photochemical
activation method21. This is unexpected, since having a large ligand could have electronic repulsion
with the NHC group on top of the catalyst and with the chloride groups on the ruthenium. The electronic repulsion would mean an increase in energy for the cis complex. This increase in energy, in turn, could causethe shift in the activation. However, all groups tested on the sulphur atom showed no sign of this possibility39. The groups tested include; phenyl, β-naphthyl, 1-pyrenyl and isopropyl39.
Thus, an attempt to lower the energy barrier by destabilising the cis complex via the increase of the side groups on the sulphur is futile.
The cis configuration is the thermodynamic complex because the trans configuration is electronically destabilised by the NHC ligand63. However, the stability of the cis complex is dependent on the
solvent39. Where polar solvents stabilise the cis configuration, a lower polarity provides diverse
results. An example of a complex where the trans configuration is nearly always the thermodynamic complex came from a study on N-chelated complexes. In this research the complexes shown in figure 14 with SIXyl and SITol as R groups on the NHC were active at room temperature in lower polarity solvents39. The authors who observed this gave no explanation for this phenomena39. Although a
trans active complex might sound beneficial it would consequently lack the property of externally activating the complex. Thus, in order to utilise the isomerisation as an activation mechanism a polar solvent is favoured.
Figure 14. Hoveyda-Grubbs type catalyst which can be trans and cis thermodynamically favoured depending on the solvent.
18
Mechanism of the activation
The mechanism for the photochemical activation starts with irradiating the chelated sulphur-ruthenium bond creating a 14 electron species21. The broken sulphur-chelated bond will rotate away
from the complex, presumably to allow the chloride to rotate in the trans position as shown in figure 15. After the rotation of the chloride the sulphur is coordinated back onto the ruthenium at the trans position. This process requires a specific wavelength to excite the sulphur-ruthenium bond. The wavelength reported to stimulate the ruthenium-sulphur bond to start the isomerisation mechanism is 350 nm58,64,7.
Figure 15. Supposed mechanism of the cis, trans isomerisation21.
It is surprising that the sulphur is coordinating back onto the metal to create an active catalyst, instead of staying dissociated, since the dissociation of the chelated atom is necessary to allow olefin coordination as explained in the initiation of the Hoveyda-Grubs catalyst. Furthermore, the 14 electron complex seems more active than the trans complex. The re-coordination of the sulphur bond is not explained in the article21. The results on the activity of the cis and trans complexes shows
clear evidence that the trans complex is the only active complex7,61. Thus, it appears that the trans
complex is active despite the sulphur coordinate back onto the ruthenium creating an 16 electron complex. Therefore the question remains why the cis complex is inactive and the trans complex is active when they both need to be initiated to become 14 electron complexes before undergoing the propagation step of the metathesis reaction.
Speculating on the activity difference between the trans and cis complex resulted in the idea that there could be a trans effect of the NHC. As the trans effect would explain the trans re-coordinated sulphur being able to dissociate where the cis coordinated would not. Since the cis ligands would be strongly bound without the destabilisation of the trans ligand. Supporting this idea are: the role which the NHC plays in the phosphorous dissociation and the mechanism for the Hoveyda-Grubbs catalyst indicating the use of a trans isomer52,34,48. The fact that the cis configuration is the
thermodynamic complex and assuming trans effect is the reason for the possibility of dissociating the sulphur in the trans configuration, the cis complex would be a pre-catalyst. This would prevent the cis configuration from being active in ambient conditions.
19
Reversion of the trans configuration
The cis configuration is the thermodynamic complex and therefore over time the trans complex will reverse back into the cis complex58. The time it takes for the trans complex to decline varies
significantly, for example: with TolSCF3 shown in figure 16 the pure cis-complex was observed after 10
hours. Yet in the decline of the Dipp-SCF3 complex, the trans-configuration was observed even after
three weeks. The suggested reason for the vast differences on the isomerisation rate are steric effects
58. Besides the size of the NHC groups influencing the life time, the size of the side groups on the
chelated sulphur will impact the isomerisation speed too. Surprisingly, the decrease in bulky side groups on the chelated sulphur will result in a slower isomerisating61.
Figure 16. Different types of the NHC ligands, which have different declining times.
The solvent effect also contributes to the lifetime of the trans complex61. The lower the polarity of
the solvent the slower the isomerisation from the cis into the trans complex; in solvent of lower polarity the rate of the shift back to the cis complex is decreased as well. For example, using benzene as a solvent the trans isomer was produced with a conversion of 85%; after 15 days 45% was converted back to the cis isomer61. With a more polar solvent like DMSO the conversion back is
nearly immediate as the trans-configuration was not detected61. Therefore, the isomerisation rate is
decreased by the nonpolar solvents, which is a stabilising effect on the life time of the trans complex.
Deactivation
Besides converting back into the cis configuration, photochemically activated complexes could also be deactivated as well. As seen in figure 15, heat is a method to deactivate the complex. This was shown with an experiment where the complex was activated for 15 minutes, followed by 5 minutes of heating at 80 degrees. In this experiment the full conversion to the cis complex was observed21. The
possibility of externally controlling the activation and deactivation of the complex could be interpreted as an on/off switch. There is, however, a more radical deactivation method for this type of complexes: the kill switch.
20
The kill switch, which will decompose the complex, is performed by irradiating at a wavelength of 254 nm and using the socalled supersilyl substituent on the NHC backbone as shown in figure 17. This supersilyl groups are proven essential as the catalyst will not be decomposed without these groups65,66. To prove the decomposition, the complex was inactivated with 254 nm followed by
activation using 350 nm. In this experiment the complex remained inactive and thus prove the decomposition67. Thus, a catalyst with a supersilyl group is able to kill the reaction. The mechanism of
this decomposition reaction has not been clarified in the article, however, the involvement of the supersilyl groups appears to be the key.
Activation by heat
Another stimulus used on the same type of catalysts to activated the complexes is heat58. The light
and thermal activation of the catalysts are comparable methods. There is only a slight difference in the mechanisms, which is the sulphur coordinating back onto the metal in the case of photochemical activation. In the thermally activated mechanism the sulphur-metal bond is excited and dissociates but does not coordinate back in the trans position as with photochemical activation. The difference in mechanisms between the two activation methods is not discussed in the articles which compare the two methods58. Despite not discussing the difference, the absence of re-coordination is an
advantage. As the thermally activated catalyst remains a 14 electron species58, which can directly
coordinate the olefin substrate onto the complex68.
The dissociation of the sulphur in the photochemically activated complex is an extra step which will increase the energy barrier compared to the thermally activated complex. As the initiation step is the rate determining step the 14 electron complex after heating is an advantage. As explained previously, the activation of the Hoveyda-Grubbs catalyst consisted of multiple steps48. The first step in the
process is the dissociation of the sulphur bond, which theoretically needs to happen at least twice for the photochemically activated catalyst. The energy barrier decrease was predicted by a DFT study done on a nitrogen chelated atom catalyst where large difference in energy between the 14 and 16 electron complex compared to the transition state were observed39. This difference is influenced by
the type of ligand. This study did not expand on sulphur chelated complexes, however, it can be assumed the results will be similar.
Figure 17 Co m pl ex de co m po sa bl e by irr ad iat io n wi th lig ht of 25 4 n m.
21
Figure 18. Difference between the 16 electron complex and the 14 electron complex with respect to the energy 41. As the
photochemically activated catalyst is an 16 electron species and the thermally activated catalyst an 14 electron species, the graph represents the difference between the light and thermally activated catalyst.
22
Reaction scope
As the olefin metathesis reaction has been around for a long time there are many alterations and applications. As previously discussed, there are multiple generations of catalysts and the catalysts are now designed for specific reactions. Therefore, the scope of the olefin metathesis reaction is immense. Since the thermally and photochemically activated catalysts are designed for the same type of substrate these will be compared. The substrates of the photochemically and thermally activated catalysts will be nearly identical as the catalyst can be identical. The only difference between most catalysts is the activation method. Consequently, there will be a focus on the general substrate scope, the difference between the photochemically and thermally activated catalyst, the steric effects, and the selectivity.
The conversion of the olefin metathesis reactions is influenced by the side group on the sulphur. In general it seems that the larger these side groups, the lower the conversion of the olefin metathesis reactant68. Two explanations have been given. First, the steric repulsion of the olefin to get towards
the coordination site in the active trans isomer. The second explanation is the reversion back to the cis isomer, leaving less trans configured isomers to accomplish the reaction. Both reasons are valid and both indicate the effect is due to the activation step. Thus, all the olefin metathesis reaction are influenced by this reduced conversion with large groups on the sulphur. Consequently, the use of small groups on sulphur seems superior as there are no disadvantages found to using small groups.
RCM
The RCM is often tested with reaction 1 in table 167, thus estimating a complete scope is complicated.
However, a general idea of the potential substrate scope would include substituted substates and substates rich in polar functional groups. The many substituents on the produced ring could be explained by the Thorpe Ingold effect, as this is beneficial for the formation.
The reaction times differ massively, from 15 minutes for most of the thermally activated and up to 10 hours for the photochemically activated catalysts. Therefore, it can be concluded that the thermal activation is superior to the photochemical action of the complex is respect to time. This is assuming the reaction time is solely dependent on the activation time. The intensity of the light used was not specified in the article58. Thus, the difference could be due to the low light intensity.
Table 1 shows that catalyst Tol-SCF3 is best in converting substrates with two methyl groups next to
the double bond, whereas the lack of the side groups on this position is more favourable for the Mes-SCF3 and DippSCF3 catalyst. This is logical as the trend of the steric bulk on the catalyst increases from
Tol-SCF3 > Mes-SCF3 > DippSCF358. Thus, it can be assumed that the increase of steric bulk of the
catalyst will improve performances towards substrates with decreased steric on the position next to the double bonds. The steric bulk on the opposite side of the coordinating double bond appears to have less influence on the conversion. The change of the two toluenesulfonyl (Ts) groups to the two CO2Et groups has a minimal effect on the conversion. The Ts does have a little better conversion
shown in reaction 4 than the two CO2Et groups shown in reaction 3. This effect is mainly noticeable in
catalyst Mes-SCF3 and Dipp-SCF3, as they do not produce any product in reaction 3 and a little in
reaction 4.
In table 1 the conversions to the products are shown. It shows the type of stimuli, the complex used, and gives an overview of the different conversions for the complexes and stimuli used. The low conversion of the photochemically activated complexes compared to the thermally activated complexes is surprising and most notable in reaction 4 using catalysts Tol-SCF3 and Mes-SCF3. The
reason was not addressed in the article58, however, the knowledge that the 14 electron complex has
a lower barrier to the transition state than the 16 electron complex could provide an insight in the reason for this difference. As discussed before the photochemically activated complex become a 16
23
electron complex again and the thermally activated complexes stays a 14 electron complex. The general trend is that the thermally activated complexes have a higher conversion compared to the photochemically activated complexes. As the barrier for the 16 electron complexes is higher this could indicate that less complexes are converted into active species. The difference could also be caused by the decline of the photochemically activated catalyst back into the cis configuration. The lower conversion for the photochemically activated catalyst makes the photochemically activated catalyst is not ideal for the RCM reaction.
Table 1. Catalyst tested for the conversion efficiency above and a table of the conversion percentages for the different complexes below. The conversion of the thermally activated complexes were heated to 80°C for 15 min, except for the a
complexes there where 80°C for 1h.
reaction Product Time
for light Photochemical activation conversion Time for heat Thermal activation conversion Source 1 30 min Tol-SCF3 58 % Mes-SCF3 90 % Dipp-SCF3 94 % 15 min Tol-SCF3 52 % Mes-SCF3 71 % Mes-SCF3 75 %a Dipp-SCF3 100 % 58 69 2 35 min Tol-SCF3 81% Mes-SCF3 100 % Dipp-SCF3 94 % 15 min Tol-SCF3 91 % Mes-SCF3 90 % Dipp-SCF3 100 % 58 3 6 hours Tol-SCF3 91 % Mes-SCF3 0 % Dipp-SCF3 0 % 58 4 10 hours Tol-SCF3 92 % Mes-SCF3 7 % Dipp-SCF3 9 % 15 min Tol-SCF3 94 % Mes-SCF3 20 % Dipp-SCF3 38 % 58
24
Besides reactions of both thermally and photochemically activated catalyst there are some examples of thermally activated reactions only. As the thermally and photochemically activated catalyst only differ in activation method, the photochemically activated catalyst could be able to accomplish this reaction too. However, as is the thermally activated catalysts reaction have a broader scope than the photochemically activated catalysts.
Table 2. . Catalyst tested for the conversion efficiency above and a table of the conversion percentages for the different complexes below. The conversion of the thermally activated complexes were heated to 80°C for 15 min, except for the a
complexes there where 80°C for 1h.
reaction Product Thermal activation conversion Source
1 Mes-SCF3 88 %a 69 2 Tol-SCF3 70 % Mes-SCF3 23 % Mes-SCF3 97 %a Dipp-SCF3 98 % 58 69 3 Tol-SCF3 90 % Mes-SCF3 34 % Mes-SCF3 77 %a Dipp-SCF3 100 % 58 69
ROMP
Similar to the RCM, the ROMP the scope contains both polar and nonpolar substrates. The cyclononene and the cyclononadiene are nonpolar substrates used in polymerisation reactions. Besides these substrates unstable rings are also known to be substrates in the ROMP6. However, the
substrate of reaction 3 in table 2 are mostly used by the photochemically activated catalyst39. As the
catalysts are mostly tested with reaction 3 the scope could be larger, just as for the RCM. The substrates for the ROMP are generally unstable rings with or without polar groups.
Table 2. Conversions for the polymerisation reaction by the photochemically activated catalysts shown at the right70.
25
Interestingly, the conversion of the nonpolar substrates is higher than the conversion of the substrate with the CO2Et groups. It is unclear if this is due to polarity or because of the steric bulk on the NHC
ligand. As the CO2Et groups are pointing away from the metal it is more likely that the steric bulk is
the reason for the difference in conversion between reaction 2 and 3 in table 2. The bridging ring could interfere with the methylene groups as shown in figure 19. The steric repulsion is proposed to be the reason for the low conversion.
Figure 19. Illustration of the supposed steric repulsion in the activated catalyst, with the substrates of the polymerisation reaction.
Selectivity
In an attempt to improve the selectivity of the general catalyst, asymmetric NHC ligands were designed and tested44. Surprisingly, the symmetric NHC ligands outperformed the asymmetric NHC
ligands shown in figure 20. The symmetric ligands favoured the Z product whereas the asymmetric produced a mixture. This, however, was tested with Hoveyda-Grubbs oxygen chelated complexes. The sulphur chelated complexes where not included in this research44. Nevertheless, in the initiation
step as explained previously, the entire chelated carbene is dissociating. Therefore, the size of the NHC in the chelated oxygen complexes is representable for the chelated sulphur complexes.
26
Potential
Polymerisation
The first and second generation olefin metathesis catalyst are not broadly used for ROMP71,72. It is
hypothesised that this is because of the relatively broad PDI values for a living polymerisation, ranging from 1.3 to 1.5 due to the rate determining initiation step73. The potential of the
photochemically activated complexes for polymerisation reactions has been recognised67. As
lowering the PDI value by exploring the decomposition or kill switch to limit the diversity in polymeric length is an possibility. Since the olefin metathesis polymerisation is a living polymerisation the PDI should be closed to the limited when: the rate of every individual catalyst is the same, the activation is at the same time, and the termination is at the same time. As the activation and termination could be externally controlled, the controllability of the PDI will increase. The assumption of a lower PDI in these conditions could only be true when the rate of all the individual catalysts is the same. When the rate determining step is also externally controlled the set-up for a low PDI value is theoretically accomplished.
The kinetics is not only defined by the rate-determining step. Although largely focused on the improvement of the rate-determining step, the kinetics of the propagation step contributes to the total rate as well, especially with a polymerisation43. In a living polymerisation reaction the
assumption is made that the active chain concentration remains constant throughout the reaction, this results in a pseudo-first-order reaction in monomer concentration for the propagation step73. To
put this into perspective, when the concentration of the monomer becomes more equalised around the individual catalysts this decreases the dependency on monomer concentration. This is with the assumption that it is a well-mixed system and the assumption that light will penetrate the solution equally. As the polymerisation reaction is dependent on the initiation of the catalyst and the concentration of the monomers, this provides a potentially viable method for polymerisation. The relatively large range of values for the PDI within the ROMP reaction might be due to a unfavourable substrate to catalyst concentration ratio. A study done on sulphur chelated catalysts shows that the yields and PDI value vary when there are different combinations of substrate and catalyst concentration6. There can be reasoned that the contribution of the substrate and catalyst
concentration ratio on the yield of the reaction is more prevalent in the ROMP reaction than in the other types of metathesis reactions. Since the contribution of the propagation step, which is mainly dependent on the concentration ratio, is more prevalent in the ROMP than in the other types of reactions. The potential to utilise the photochemically activated ROMP reaction for the advantages described above further research is needed. For example looking into the optimal concentration ratio for the specific catalysts with the capability of the on/off switch.
Sustainability
The environment is an important topic at the moment, consideration of the environmental impact of a reaction has become increasingly important74,75. Green chemistry is built upon 12 principles76.
Design for Energy Efficiency is the main reason the photochemical activation might be superior. The principle of design for energy efficiency states that energy requirements should be minimized. This can, for example, be achieved by operating at ambient temperature and pressure.
The energy efficiency of the process could be increased by photochemically activating the catalyst with sunlight. The light used to excite the ruthenium-chelated bond is 350 nm which falls into the ultra violet range77. This is included in the spectrum of sunlight which ranges from 290 to 720 nm78.
Although, the activation by sunlight would be beneficial, there are a few drawbacks to the proposal. For example, the possible activation of the kill switch, low intensity, and lack of sunlight.
27
When the kill switch is incorporated into the system, it ought not be activated by sunlight. Since, activating directly following by killing the catalyst due to sunlight would be a waste of energy and materials. Fortunately, the activation wavelength of 254 nm is not significantly present in sunlight. Thus, an external light source is vital to activate the kill switch when desired. It should be taken into consideration that such an external light source will decrease the energy efficiency.
In order to excite the ruthenium-chelated bond the intensity of the light should be sufficient. Unfortunately, the intensity of the light required to excite the bond is not discussed in the articles used in this paper7,21,58. However, the intensity of the 350 nm in sunlight is estimated to be low and
therefore inadequate to activate the complex. The estimation is based on figure 21. To overcome the low intensity, lenses and prisms should be deployed.
Figure 21. The intensity of light at different temperatures79.
During the night there is a lack of sunlight which intervenes with the activation of the catalyst. While this is true, a continuous activation is not necessary when the trans configured form of the catalyst has an extensive life time. As the life span of a trans configuration can be up to a few weeks61, the
delay of reactivation could increase with the same amount. Thus, continuous stimuli is unnecessary. However, a regular reactivation would stabilise the concentration of the active catalysts.
The higher probability for sustainable photochemically activated catalyst is an advantage over lower probability for sustainable thermally activated catalyst, which is explored above. Nevertheless, the energy efficiency of the thermal activated catalyst could increase by the use of sustainable energy sources. For example, the use of solar panels to generate energy. Still, the energy wasted by the solar panels is less favourable than directly activating the complex with sunlight.
28
Conclusion
The thermally activating method appears to be superior to the photochemically activating method in nearly every way. The scope appears to be larger and the conversion is higher. This is presumably due to the lower energy barrier of the 14 electron thermally activated complex to the transition state compared to the higher energy barrier of the 16 electron photochemically activated complex to the transition state. Since, the active form of the thermally activated complex is a 14 electron complex and the photochemically activated complex is a 16 electron complex. Besides the lower energy barrier the time the thermally activated complexes are heated is less than the excitation time for the photochemically activated complexes.
For both activation mechanisms there are mainly steric effects which influence the conversion of the reactions within the reaction scope. The increase of steric bulk on the NHC backbone and the decrease of steric bulk on the substrate or the inversed combination is optimal for the conversion. Large groups on the sulphur atom are decreasing the conversion. Thus the size of the substrate determines the ideal steric bulk of the NHC back bone and large groups on the sulphur atom are undesirable.
The photochemically activated complexes do have potential in polymerisation reactions, since the photochemically activated complexes have an on/off or even a kill switch. This gives potential for obtaining lower PDI values. The photochemically activated catalyst also has the potential to become a fairly green reaction, since it could be activated with sunlight. The success of the sunlight photochemical activation has to be further investigated because the intensity of the 350 nm in sunlight is low. However, there certainly is a potential for this sunlight photochemical activation. Thus, the thermally activated catalyst is superior to the photochemical activation with exception of the potential towards the green aspect and the polymerisation reaction.
29
References
1. Mai S, González L. Molecular Photochemistry: Recent Developments in Theory. Angew Chemie
Int Ed. June 2020:anie.201916381.
2. Powles SB. Photoinhibition of Photosynthesis Induced by Visible Light. Annu Rev Plant Physiol. 1984;35(1):15-44.
3. Yachandra VK, Sauer K, Klein MP. Manganese cluster in photosynthesis: Where plants oxidize water to dioxygen. Chem Rev. 1996;96(7):2927-2950.
4. The Nobel Prize in Chemistry 2005.
https://www.nobelprize.org/prizes/chemistry/2005/summary/. Accessed April 2, 2020.
5. Tzur E, Szadkowska A, Ben-Asuly A, et al. Studies on Electronic Effects in O-, N- and S-Chelated Ruthenium Olefin-Metathesis Catalysts. Chem - A Eur J. 2010;16(29):8726-8737.
6. Diesendruck CE, Vidavsky Y, Ben-Asuly A, Lemcoff NG. A latent s-chelated ruthenium benzylidene initiator for ring-opening metathesis polymerization. J Polym Sci Part A Polym
Chem. 2009;47(16):4209-4213.
7. Levin E, Mavila S, Eivgi O, Tzur E, Lemcoff NG. Regioselective Chromatic Orthogonality with Light-Activated Metathesis Catalysts. Angew Chemie Int Ed. 2015;54(42):12384-12388.
8. Nguyen SBT, Johnson LK, Grubbs RH, Ziller JW. Ring-Opening Metathesis Polymerization (ROMP) of Norbornene by a Group VIII Carbene Complex in Protic Media. J Am Chem Soc. 1992;114(10):3974-3975.
9. Phillips JH. Latest Industrial Uses of Olefin Metathesis. In: Organometallic Chemistry in
Industry. Wiley; 2020:259-282.
10. Farina V, Shu C, Zeng X, et al. Second-generation process for the hcv protease inhibitor biln 2061: A greener approach to ru-catalyzed ring-closing metathesis. Org Process Res Dev. 2009;13(2):250-254.
11. Rosenquist Å, Samuelsson B, Johansson PO, et al. Discovery and development of simeprevir (TMC435), a HCV NS3/4A protease inhibitor. J Med Chem. 2014;57(5):1673-1693.
12. Kong J, Chen CY, Balsells-Padros J, et al. Synthesis of the HCV protease inhibitor Vaniprevir (MK-7009) using ring-closing metathesis strategy. J Org Chem. 2012;77(8):3820-3828.
13. Van Leeuwen PWNM. Homogeneous Catalysis Understanding the Art.; 2004. 14. Crabtree RH. The Organometallic Chemistry of the Transition Metals.
15. Rybtchinski B, Milstein D. Metal insertion into C-C bonds in solution. Angew Chemie - Int Ed. 1999;38(7):870-883.
16. Grubbs RH. Olefin metathesis. In: Tetrahedron. Vol 60. Pergamon; 2004:7117-7140.
17. Gordon CP, Copéret C. Metal Alkyls with Alkylidynic Metal‐Carbon Bond Character: Key Electronic Structures in Alkane Metathesis Precatalysts. Angew Chemie Int Ed. March 2020:anie.201915557.
18. Schuster M, Blechert S. Olefin Metathesis in Organic Chemistry. Angew Chemie Int Ed English. 1997;36(19):2036-2056.
30
19. Clayden J, Greeves N, Warren SG. Organic Chemistry.
https://global.oup.com/ukhe/product/organic-chemistry-9780199270293?cc=nl&lang=en&. Accessed November 13, 2018.
20. Navratilova I, Skladal P, Viklicky V, et al. Organic Chemistry 2nd Edition. Vol 55.; 2012.
21. Ben-Asuly A, Aharoni A, Diesendruck CE, et al. Photoactivation of ruthenium olefin metathesis initiators. Organometallics. 2009;28(16):4652-4655.
22. Braddock DC, Wildsmith AJ. On the use of tandem allylic acetate isomerisation and ring-closing metathesis with palladium(0) phosphine complexes and ruthenium benzylidenes as orthogonal catalysts. Tetrahedron Lett. 2001;42(18):3239-3242.
23. Olefin Metathesis and Metathesis Polymerization - K. J. Ivin, J. C. Mol - Google Boeken. Accessed April 2, 2020.
24. Liu F, Xu N, Ling L, Hu J, Zhang H. Regio- and stereoselective ring-opening metathesis polymerization of 3-ferrocenyl substituted cyclooctenes and copolymerization with norbornene derivatives. Eur Polym J. 2020;124:109472.
25. Lynde BE, Maust RL, Li P, Lee DC, Jasti R, Boydston AJ. Ring-opening metathesis polymerization of a strained stilbene-based macrocyclic monomer. Mater Chem Front. 2020;4(1):252-256. 26. Steed JW, Tocher DA, Rogers RD. Ruthenium-mediated cyclodimerisation of buta-1,3-diene.
Chem Commun. 1996;(13):1589-1590.
27. Kensy VK, Tritt RL, Haque FM, et al. Molecular Weight Control via Cross Metathesis in Photo-Redox Mediated Ring-Opening Metathesis Polymerization. Angew Chemie Int Ed. March 2020. 28. Bielawski CW, Grubbs RH. Living ring-opening metathesis polymerization. Prog Polym Sci.
2007;32(1):1-29.
29. Matyjaszewski K, Xia J. Atom transfer radical polymerization. Chem Rev. 2001;101(9):2921-2990.
30. Schwab P, Grubbs RH, Ziller JW. Synthesis and applications of RuCl2(=CHR′)(PR3)2: The influence of the alkylidene moiety on metathesis activity. J Am Chem Soc. 1996;118(1):100-108.
31. Zaranek M, Robaszkiewicz J, Janica I, Gauvin RM, Pawluć P, Mortreux A. In situ Mo(CO)₆-based catalysts for alkyne metathesis: Silanols vs phenols as co-catalysts under thermal and photochemical activation. Catal Commun. 2020;138:105944.
32. Benedikter MJ, Ziegler F, Groos J, Hauser PM, Schowner R, Buchmeiser MR. Group 6 metal alkylidene and alkylidyne N-heterocyclic carbene complexes for olefin and alkyne metathesis.
Coord Chem Rev. 2020;415:213315.
33. Song J-A, Peterson GI, Bang K-T, et al. Ru-Catalyzed, cis -Selective Living Ring-Opening Metathesis Polymerization of Various Monomers, Including a Dendronized Macromonomer, and Implications to Enhanced Shear Stability. J Am Chem Soc. May 2020:jacs.0c02785.
34. Fürstner A. Olefin Metathesis and Beyond. Angew Chemie Int Ed. 2000;39(17):3012-3043. 35. Straub BF. Ligand Influence on Metathesis Activity of Ruthenium Carbene Catalysts: A DFT
31
36. Scholl M, Ding S, Lee CW, Grubbs RH. Synthesis and activity of a new generation of ruthenium-based olefin metathesis catalysts coordinated with 1,3-dimesityl-4,5-dihydroimidazol-2-ylidene ligands. Org Lett. 1999;1(6):953-956.
37. Kingsbury JS, Harrity JPA, Bonitatebus PJ, Hoveyda AH. A recyclable Ru-based metathesis catalyst. J Am Chem Soc. 1999;121(4):791-799.
38. Garber SB, Kingsbury JS, Gray BL, Hoveyda AH. Efficient and recyclable monomeric and dendritic Ru-based metathesis catalysts. J Am Chem Soc. 2000;122(34):8168-8179.
39. Pump E, Leitgeb A, Kozłowska A, et al. Variation of the Sterical Properties of the N-Heterocyclic Carbene Coligand in Thermally Triggerable Ruthenium-Based Olefin Metathesis Precatalysts/Initiators. Organometallics. 2015;34(22):5383-5392.
40. Opstal T, Verpoort F. Synthesis of Highly Active Ruthenium Indenylidene Complexes for Atom-Transfer Radical Polymerization and Ring-Opening-Metathesis Polymerization. Angew Chemie
Int Ed. 2003;42(25):2876-2879.
41. Savka R, Foro S, Gallei M, Rehahn M, Plenio H. Oxidation-Triggered Ring-Opening Metathesis Polymerization. Chem - A Eur J. 2013;19(32):10655-10662.
42. Zirngast M, Pump E, Leitgeb A, Albering JH, Slugovc C. Pyridine as trigger for chloride isomerisation in chelated ruthenium benzylidene complexes: Implications for olefin metathesis. Chem Commun. 2011;47(8):2261-2263.
43. Sandford MS, Love JA, Grubbs RH. Mechanism and activity of ruthenium olefin metathesis catalysts. J Am Chem Soc. 2001;123(27):6543-6554.
44. Ledoux N, Linden A, Allaert B, Mierde H Vander, Verpoort F. Comparative Investigation of Hoveyda–Grubbs Catalysts bearing Modified N-Heterocyclic Carbene Ligands. Adv Synth Catal. 2007;349(10):1692-1700.
45. Sanford MS, Ulman M, Grubbs RH. New insights into the mechanism of ruthenium-catalyzed olefin metathesis reactions [9]. J Am Chem Soc. 2001;123(4):749-750.
46. Burdett KA, Harris LD, Margl P, et al. Renewable monomer feedstocks via olefin metathesis: Fundamental mechanistic studies of methyl oleate ethenolysis with the first-generation grubbs catalyst. Organometallics. 2004;23(9):2027-2047.
47. Dwyer CL, Kirk MM, Meyer WH, Van Rensburg WJ, Forman GS. Rotational isomerism of a phoban-derived first-generation grubbs catalyst. Organometallics. 2006;25(16):3806-3812. 48. Solans-Monfort X, Pleixats R, Sodupe M. DFT Mechanistic Study on Diene Metathesis Catalyzed
by Ru-Based Grubbs-Hoveyda-Type Carbenes: The Key Role of π-Electron Density Delocalization in the Hoveyda Ligand. Chem - A Eur J. 2010;16(24):7331-7343.
49. Quintin F, Pinaud J, Lamaty F, Bantreil X. Mechanosynthesis of Noels-type NHC-Ruthenium Complexes and Applications in Ring-Opening Metathesis Polymerization. Organometallics. 2020;39(5):636-639.
50. Sivanesan D, Seo B, Lim CS, Choi D, Kim T, Kim HG. An additional potential donor-bearing alkylidene-containing latent NHC-ruthenium-based catalyst for olefin metathesis polymerization. Polym Chem. 2020;11(14):2511-2518.
51. Torker S, Merki D, Chen P. Gas-phase thermochemistry of ruthenium carbene metathesis catalysts. J Am Chem Soc. 2008;130(14):4808-4814.
32
52. Lappert MF. COORDINATION CHEMISTRY OF BNALENT GROUP IV DONORS;
NUCLEOPHILIC-CARBENE AND DIALKYLSTANNYL ENE COMPLEXES. Vol 100.; 1975.
53. Herrmann WA, Köcher C. N-Heterocyclic Carbenes. Angew Chemie Int Ed English. 1997;36(20):2162-2187.
54. Urbina-Blanco CA, Leitgeb A, Slugovc C, et al. Olefin Metathesis Featuring Ruthenium Indenylidene Complexes with a Sterically Demanding NHC Ligand. Chem - A Eur J. 2011;17(18):5045-5053.
55. Tzur E, Szadkowska A, Ben-Asuly A, et al. Studies on Electronic Effects in O-, N- and S-Chelated Ruthenium Olefin-Metathesis Catalysts. Chem - A Eur J. 2010;16(29):8726-8737.
56. Courchay FC, Sworen JC, Wagener KB. Metathesis activity and stability of new generation ruthenium polymerization catalysts. Macromolecules. 2003;36(22):8231-8239.
57. Stewart IC, Ung T, Pletnev AA, Berlin JM, Grubbs RH, Schrodi Y. Highly efficient ruthenium catalysts for the formation of tetrasubstituted olefins via Ring-Closing Metathesis. Org Lett. 2007;9(8):1589-1592.
58. Ivry E, Frenklah A, Ginzburg Y, et al. Light-and Thermal-Activated Olefin Metathesis of Hindered Substrates. Organometallics. 2018;37(2):176-181.
59. Theunissen C, Ashley MA, Rovis T. Visible-Light-Controlled Ruthenium-Catalyzed Olefin Metathesis. J Am Chem Soc. 2019;141(17):6791-6796.
60. Barbasiewicz M, Szadkowska A, Bujok R, Grela K. Structure and activity peculiarities of ruthenium quinoline and quinoxaline complexes: Novel metathesis catalysts. Organometallics. 2006;25(15):3599-3604.
61. Aharoni A, Vidavsky Y, Diesendruck CE, Ben-Asuly A, Goldberg I, Lemcoff NG. Ligand isomerization in sulfur-chelated ruthenium benzylidenes. Organometallics. 2011;30(6):1607-1615.
62. Atkins P, Overton T. 5th Edition_Inorganic Chemistry by Duward Shriver. Oxford University Press, USA; 2010.
63. Benitez D, Goddard WA. The isomerization equilibrium between cis and trans chloride ruthenium olefin metathesis catalysts from quantum mechanics calculations. J Am Chem Soc. 2005;127(35):12218-12219.
64. Weitekamp RA, Atwater HA, Grubbs RH. Photolithographic olefin metathesis polymerization. J
Am Chem Soc. 2013;135(45):16817-16820.
65. Dinger MB, Nieczypor P, Mol JC. Adamantyl-Substituted N-Heterocyclic Carbene Ligands in Second-Generation Grubbs-Type Metathesis Catalysts. Organometallics. 2003;22(25):5291-5296.
66. Herbert MB, Suslick BA, Liu P, et al. Cyclometalated Z-Selective Ruthenium Metathesis Catalysts with Modified N-Chelating Groups. Organometallics. 2015;34(12):2858-2869.
67. Sutar RL, Levin E, Butilkov D, Goldberg I, Reany O, Lemcoff NG. A Light-Activated Olefin Metathesis Catalyst Equipped with a Chromatic Orthogonal Self-Destruct Function. Angew
Chemie Int Ed. 2016;55(2):764-767.
68. Thiel V, Hendann M, Wannowius KJ, Plenio H. On the mechanism of the initiation reaction in Grubbs-Hoveyda complexes. J Am Chem Soc. 2012;134(2):1104-1114.
