Design of new aggregates for catalysis
Tosi, Filippo
DOI:
10.33612/diss.107814277
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2019
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Tosi, F. (2019). Design of new aggregates for catalysis. Rijksuniversiteit Groningen. https://doi.org/10.33612/diss.107814277
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Chapter 5
Towards New Catalytically Active
Amphiphiles
ABSTRACT: The two sections of this chapter focus on the synthesis and application of homochiral amphiphiles for catalysis in self-assembled bilayers. The first section reports the synthesis of new BINOL-based amphiphiles for catalysis and attempts to perform self-assembly. The second section is centered on the catalytic application of salen-based amphiphiles described in Chapter 3.
80
5
Section I – BINOL-Based Amphiphiles
for Catalysis
5.1.1 Introduction
Chiral phosphoric acids are well-known organocatalysts which have been used for stereoselective transformations under mild conditions.[1–3] A typical feature of
these organocatalysts is their binding pocket in which they can accommodate a substrate by hydrogen bonding (Figure 5.1.1). The high tunability of the skeleton scaffold has made them particularly interesting systems for catalytic application. The angle between the groups forming the chiral pocket can be tuned to give a different geometry and the electronics or steric properties can be changed easily by modifying the substituents in the positions adjacent to the phosphoric acid moiety.[4]
The renaissance of such systems began in 2004 with the design and application of BINOL-based phosphoric acids by Akyiama and co-workers,[5] which set the stage
for several new generations of scaffolds that have been used in stereoselective catalysis.[6–10]
The activation mode of phosphoric acids is highly dependent on the pKa of the
acid itself, which coordinates an electrophilic substrate by protonation, lowering its LUMO and thus making it more prone to nucleophilic attack.[11] It is important to
consider that there is not a positive linear correlation between the pKa of the acid and
the stereoselectivity, in fact, the process works very much the opposite. Strong acids do not bind the substrate effectively in the chiral pocket, resulting in too strong activation to effectively influence the stereoselective outcome of the catalysis. An interesting study carried out jointly by the groups of Rueping and Gschwind has pointed out the different species arising by coordination of the substrates, highlighting the double contribution of ion pairing and hydrogen bonding as modes of activation for organic phosphoric acid catalysts.[12]
Figure 5.1.1 General features of a chiral phosphoric acid.
The presence of aromatic substituents forming the binding pocket favors also the inclusion of organic substrates making chiral phosphoric acids effective systems for supramolecular chemistry as well. Recently our group has reported the successful
81
5
application of such systems in enantioselective liquid-liquid extraction (ELLE) due to their efficient binding ability to amines.[13–15] They have been applied in more complex
supramolecular systems, including catenanes, which have been used for anion binding[16] as well as for catalysis.[17] Interesting feature of the above-mentioned
catenanes is that, due to their interlocked structure, reagents are brought close together in a chiral environment and are double activated by the chiral phosphoric acids. The encapsulation and subsequent double activation of the substrates determines the outcome of the catalytic transformation, showing a net increase in e.e. when the catenane-catalyst is used.[17]
Considering the well-known ability of chiral phosphoric acids to favor double substrate activation, we decided to implement them to perform catalysis in confined space, in our case a self-assembled bilayer. We designed a class of amphiphilic chiral phosphoric acids which feature dodecyl tails and polyethylene glycol (PEG) tails as hydrophobic and hydrophilic components, respectively. This section of the chapter describes the synthesis of this new class of phosphoric acids and the self-assembly studies in water.
5.1.2 Design
Our design for amphiphilic BINOLs and phosphoric acids for catalysis was inspired by the results showed in Chapter 2. Having already synthesized an amphiphilic BINOL which was able to self-assemble in tightly packed stable bilayers and being able to tune its self-assembly resulting in a variety of nanostructures, this design was an effective starting point for our investigation. Our first design is the reversed version of the nanotube forming surfactant reported in Chapter 2 (Figure 5.1.2).
Figure 5.1.2: Design of amphiphilic BINOL 2 and comparison with 1 (Chapter 2).
Since the aim of the project was to perform catalysis in the self-assembled bilayer, in which organic substrates should be solubilized, we considered that it was crucial to reverse the substitution onto the BINOL core in comparison with our design of amphiphile 1 presented in Chapter 2. We initially decided not to change the characteristics of the chains, and therefore focus on the synthesis compound 2 (Figure 5.1.2).
Since the self-assembly behavior of new surfactants cannot be easily predicted, we planned to synthesize similar amphiphiles featuring a difference in chain termination. A second design for new BINOL-based surfactants featured a slight
82
5
difference in the hydrophilic residue, by adding a carbon atom as linker between the aromatic core of the catalyst and the PEG chains attached (3). A final design we sought to explore was a modified version of 3 in which the terminality of the PEG chain was changed to a hydroxyl group (i.e. amphiphile 4) (Scheme 5.1.1).
Scheme 5.1.1: Design of new amphiphilic BINOLs 3 and 4.
5.1.3 Synthesis
The synthesis of our first design was performed starting from commercially available (S,S)-BINOL 5, which was protected as a methoxy-methyl ether (MOM) in quantitative yield (Scheme 5.1.2). The MOM function was meant to introduce an easily removable protecting group, which at the same time can act as an efficient ortho-directing group for functionalization at the 3,3’-position on the BINOL core. The alkyl chain was attached by double ortho-lithiation with t-BuLi, followed by quenching with 1-iodododecane, affording compound 7 in 77% yield. Attempts to substitute compound 7 in the 6,6’-position with two bromine atoms resulted in a mixture of products. A more selective synthesis route was chosen performing di-bromination of compound 8 obtained by treatment of compound 7 with aq. HCl in nearly quantitative yield.
83
5
Di-bromination of compound 8 at the 6,6’-positions on the BINOL core was performed with elemental bromine affording compound 9 in 97% yield. Subsequently, intermediate 9 was MOM protected before performing a sequence of lithiation-borylation and oxidation which allowed us to obtain compound 11, containing two hydroxyl groups necessary for PEG alkylation at the 6,6’-position.
PEG attachment to obtain compound 12 was successfully performed with our already established procedure described in Chapter 2. Unfortunately, the final MOM-deprotection step did not provide the expected result, as compound 2 proved to be unstable in air. Over a short period of time (1-2 h), the compound turned from pale yellow to dark brown just by air exposure. Column chromatography did not result in the isolation of the final compound. Repeating the reaction under milder conditions (rt) and screening different solvents (THF, MeOH) was not successful and the desired compound 2 showed instability in all cases.
Observing instability with the first designed surfactant, we reasoned that a possible explanation could be the change in electronic characteristics of the system (compared with compound 1). The change in position of the oxygen or carbon in the substitution on the aromatic system (6,6’- and 3,3’-position, respectively) apparently favors oxidation of the BINOL phenolic moieties in air making it therefore more unstable. With the intention to overcome this problem, we reasoned that inserting an additional carbon atom between the PEG chains and the aryl system would result in a more stable surfactant (compounds 3 and 4, Scheme 5.1.1). The synthesis was performed starting from compound 10 which was formylated at the 6,6’-position by lithium-halogen exchange followed by quenching with DMF (Scheme 5.1.3). Compound 13 was subsequently reduced to the benzylic alcohol 14 by treatment with NaBH4.
84
5
An attempt to improve the yield of 14 was performed by quenching the bis-lithium species obtained from compound 10 with paraformaldehyde, which allowed for the synthesis of the bis-alcohol in overall higher yield (85%) and reducing the synthetic sequence by one step.
Due to the lower acidity of the benzylic alcohol groups in compound 14 with respect to the phenol moieties in compound 11, the PEG functionalization proved to be more challenging. Under optimized conditions, we managed to successfully obtain compound 15 by deprotonation with sodium hydride and subsequent alkylation with the PEG-tosylate chains. Compound 15 was subsequently MOM-deprotected under acidic conditions affording the amphiphilic BINOL 3. Treatment with POCl3 and
subsequent hydrolysis provided the corresponding amphiphilic phosphoric acid 17, the structure of which was confirmed by 1H , 31P NMR and HRMS.
Compound 3 (compared to 2) featured two additional carbon atoms, which increase the hydrophobic character of the surfactant possibly causing difficulties or differences in its self-assembly behavior. For the synthesis of a more hydrophilic surfactant, we designed the OH terminal version of 3, i.e. 4 (Scheme 5.1.4). The synthesis of the latter required using a protected PEG chain tosylate, which would eventually result in the free OH terminality upon deprotection. Screening of similar reaction conditions for the previously described PEG attachment did not provide the desired product, and we did not manage to obtain compound 18 as precursor of our amphiphile 4. Additional attempts to synthesize the desired surfactant were performed, by using the BINOL component as the electrophile and the PEG chain as nucleophile. Tosylation reaction to obtain compound 19 and Appel reaction to obtain 20 were not successful, resulting in both cases to the deterioration/decomposition of the starting material (Scheme 5.1.4).
85
5
5.1.4 Attempts to Perform Self-Assembly
With compound 3 and its corresponding phosphoric acid 17 in hand we tried to perform self-assembly aiming at application of this novel homochiral soft-material for asymmetric catalysis. Trying to dissolve compounds 3 and 17 in water from a dried film of CH2Cl2 or CHCl3 did not prove successful, as we observed in both cases the
formation of a precipitate. Addition of THF as a co-solvent also did not improve the solubility of the surfactant. With the hope to push the self-assembly in water we tried different forms of stimulation, such as the use of sonication, heating, freezing, but none of the performed attempts proved to be effective. Using Cryo-TEM we were not able to observe any self-assembled structure.
As mentioned previously, in comparison with the BINOL-derived amphiphiles described in Chapter 2 the hydrophobic character of 3 (and 17) is higher due to the presence of the additional benzylic carbon atoms. This substitution was effective in producing a more stable compound that was easier to isolate, in comparison with our first design 2. Unfortunately, the additional carbon atoms, which make the surfactant more hydrophobic, do not allow the surfactant to be effectively solubilized in water and represented an obstacle in the completion of the project.
Attempts to favor self-assembly by addition of DOPC as a co-surfactant resulted in positive results in previous studies in our group,[18–20] but in the present case it
could not favor the self-assembly process and no aggregated structure was observed by Cryo-TEM.
5.1.5 Conclusions
In this section of the chapter, we showed the design and synthesis of a new potential amphiphile for catalysis in self-assembled soft materials. The synthesis of the second target structure designed for this BINOL-derived surfactant and the corresponding phosphoric acid was successful. Possibly due to an increased hydrophobic character of compounds 3 and 17, self-assembly of the amphiphiles was not observed.
5.1.6 Contributions
The project was carried out under the supervision of prof. dr. B. L. Feringa. The synthesis was carried out by F. Tosi. Cryo-TEM studies were performed by M. C. A. Stuart.
5.1.7 Experimental Section
Methods. Sample Preparation: 1 mL of a 2 mM solution of the compound (2, 3 and 17) in technical grade CHCl3 was placed in a 4 mL vial. The solvent (CH2Cl2 or
86
5
CHCl3) was slowly dried in a nitrogen flow and a thin film of the amphiphilic
compound was formed, which was subsequently hydrated with double distilled water.
5.1.8 Synthetic Procedures
Compound 6[21] and p-toluenesulfonate tetraethylene glycol monomethyl
ether[22] were prepared according to known literature procedures and structural data
were in accordance with the literature.
Compound 6 (500 mg, 1.33 mmol) was dissolved in dry THF (25 mL) under a nitrogen atmosphere and the mixture was cooled to -78 °C. A solution of t-BuLi (3 mL, 3.99 mmol, 1.3 M in Pentane) was added dropwise and the mixture was left stirring at 0 °C for 2 h, during which the formation of a precipitate occurred. The mixture was cooled down to -78°C and a solution of 1-iodododecane (0.984 mL, 3.99 mmol) in 5 mL of dry THF was added dropwise. The reaction mixture was slowly allowed to reach rt over the course of 16 h, quenched with sat. aq. NH4Cl,
extracted with EtOAc, dried over MgSO4, and concentrated in vacuo. Purification by
column chromatography (SiO2, Pentane → Pentane/EtOAc: 98:2) gave 7 (724.4 mg,
77%) as a colorless oil; (Rf: 0.8 Pentane/EtOAc 98:2); 1H NMR (400 MHz, CDCl3) δ 7.82
(d, J = 8.2 Hz, 2H), 7.79 (s, 2H), 7.37 (dd, J = 8.1, 6.6 Hz, 2H), 7.19 (dd, J = 7.9, 6.6 Hz, 2H), 7.14 (d, J = 8.4 Hz, 2H), 4.57 (d, J = 5.7 Hz, 2H), 4.42 (d, J = 5.7 Hz, 2H), 2.99 (m, 2H), 2.88 (m, 8H), 1.90 – 1.73 (m, 4H), 1.49 – 1.20 (m, 36H), 0.89 (t, J = 6.8 Hz, 6H); HRMS (ESI-ion trap) m/z: [M+H]+ Calcd for C48H71O4 710.5274, found 710.5244.
Compound 7 (720 mg, 1.01 mmol) was dissolved in dry 1,4-dioxane (24 mL). The reaction mixture was treated with aq. HCl (37%, 3.6 mL) and heated to 80 °C for 22 h. The mixture was cooled down to rt and concentrated in vacuo. The crude product was dissolved in CH2Cl2 (40 mL), washed with sat. aq. NaHCO3 (30 mL), dried over
MgSO4 and concentrated in vacuo. Purification by column
chromatography (SiO2, Pentane/CH2Cl2 9:1) gave 8 (586.5 mg, 93%) as a white solid;
(Rf: 0.35 Pentane/CH2Cl2 9:1); 1H NMR (400 MHz, CDCl3) δ 7.85 (d, J = 8.1 Hz, 2H), 7.82
(s, 2H), 7.35 (t, J = 7.4 Hz, 2H), 7.26 (s, 2H), 7.09 (d, J = 8.4 Hz, 2H), 5.14 (s, 2H), 2.89 (t,
J = 7.7 Hz, 4H), 1.80 (m, 4H), 1.49 – 1.22 (m, 36H), 0.91 (t, J = 6.9, 6H); 13C NMR (101
MHz, CDCl3) δ 152.2, 132.4, 132.0, 131.9, 130.2, 129.8, 128.0, 126.7, 124.3, 124.1, 32.3,
32.3, 31.2, 30.1, 30.0, 30.0, 29.9, 29.9, 29.7, 29.7, 23.0, 14.5; HRMS (ESI-ion trap) m/z: [M+H]+ Calcd for C44H63O2 623.4828, found 623.4841.
Compound 8 (580 mg, 0.93 mmol) was dissolved in dry CH2Cl2
(18 mL) under a nitrogen atmosphere and cooled down to -78 °C. Elemental bromine (0.131 mL, 2.55 mmol) was added dropwise at -78 °C, the reaction mixture was left stirring at -78 °C for 2 h, then slowly allowed to reach rt over the course of 16
87
5
h. The reaction was quenched with sat. aq. Na2S2O3, the mixture was extracted with
CH2Cl2, dried over MgSO4 and concentrated in vacuo. Purification by column
chromatography (SiO2, Pentane/CH2Cl2 9:1) gave 9 (704 mg, 97%) as a yellow oil; (Rf:
0.3 Pentane/CH2Cl2 9:1); 1H NMR (400 MHz, CDCl3) δ 7.99 (d, J = 2.0 Hz, 2H), 7.71 (s,
2H), 7.30 (dd, J = 2.0 Hz, 8.9 Hz, 2H), 6.88 (d, J = 9.0 Hz, 2H), 5.07 (s, 2H), 2.86 (t, J = 7.7 Hz, 4H), 1.75 (p, J = 7.7 Hz, 4H), 1.46 – 1.18 (m, 36H), 0.88 (t, J = 6.6 Hz, 6H); 13C NMR
(101 MHz, CDCl3) δ 152.4, 133.3, 130.9, 130.7, 130.1, 130.1, 129.4, 125.9, 118.1, 110.6,
32.2, 31.1, 30.0, 30.0, 30.0, 30.0, 29.9, 29.8, 29.8, 29.7, 23.0, 14.5; HRMS (ESI-ion trap)
m/z: [M+H]+ Calcd for C44H61Br2O2 779.3038, found 779.3067.
NaH (60%, 79.2 mg, 1.98 mmol) was suspended in dry THF (3 mL) under a nitrogen atmosphere. A solution of compound 9 (700 mg, 0.9 mmol) in dry THF (6 mL) was added dropwise to the reaction mixture at 0 °C, left stirring at 0 °C for 2 h and subsequently warmed up to rt for 30 min. The reaction mixture was then cooled to 0 °C, MOMCl (163 µL, 2.1 mmol) was added dropwise and the reaction mixture was allowed to slowly reach rt over the course of 16 h. The reaction was quenched with sat. aq. NH4Cl,
the mixture was extracted with EtOAc, dried over MgSO4 and concentrated in vacuo.
Purification by column chromatography (SiO2, Pentane/CH2Cl2 9:1) gave 10 (547.3
mg, 70%) as a yellow oil (Rf: 0.2 Pentane/CH2Cl2 9:1); 1H NMR (400 MHz, CDCl3) δ 7.98
(d, J = 2.0 Hz, 2H), 7.70 (s, 2H), 7.26 (dd, J = 2.0 Hz, 9.0 Hz, 2H), 6.96 (d, J = 9.0 Hz, 2H), 4.54 (d, J = 5.8 Hz, 2H), 4.45 (d, J = 5.8 Hz, 2H), 2.94 (m, 2H), 2.89 – 2.76 (m, 8H), 1.77 (tq, J = 13.6, 7.2 Hz, 4H), 1.48 – 1.19 (m, 36H), 0.88 (t, J = 6.7 Hz, 6H); 13C NMR (101
MHz, CDCl3) δ 153.9, 138.0, 132.3, 131.5, 129.7, 129.3, 128.3, 128.1, 125.5, 119.2, 99.2,
56.8, 32.2, 31.2, 30.6, 30.1, 30.0, 30.0, 30.0 (2C), 29.9, 29.7, 23.0, 14.5.
Compound 10 (160 mg, 0.183 mmol) was dissolved in dry THF (12 mL) and cooled to -78 °C under a nitrogen atmosphere. A solution of t-BuLi (0.633 mL, 0.823 mmol, 1.3 M in Pentane) was added dropwise and subsequently the reaction mixture was left stirring at -78 °C for 1 h, during which the color changed from yellow to dark brown. Freshly distilled B(On-Bu)3 (0.345 mL, 1.281 mmol) was added dropwise causing the
solution to turn colorless. The reaction mixture was slowly allowed to reach rt over the course of 16 h. The solvent was then removed under reduced pressure, dry toluene (12 mL) was added, followed by aq. H2O2 (0.102 mL) and the mixture was
heated at reflux for 1 h under inert atmosphere. The reaction mixture was cooled down to rt, quenched with sat. aq. NH4Cl, extracted with EtOAc, dried over MgSO4 and
concentrated in vacuo. Purification by column chromatography (SiO2, Pentane/EtOAc
8:2) gave 11 (65 mg, 48%) as a colorless oil; (Rf: 0.5 Pentane/EtOAc 8:2); 1H NMR
(400 MHz, CDCl3) δ 7.55 (s, 2H), 7.01 (s, 2H), 6.92 (d, J = 9.0 Hz, 2H), 6.66 (d, J = 9.0,
2H), 5.50 (br, 2H), 4.56 (d, J = 5.8 Hz, 2H), 4.42 (d, J = 5.8 Hz, 2H), 2.99 – 2.87 (m, 8H), 2.80 (m, 2H), 1.78 (m, 4H), 1.28 (m, 36H), 0.88 (t, J = 6.6 Hz, 6H); 13C NMR (101 MHz,
88
5
CDCl3) δ 153.0, 151.5, 137.1, 132.4, 128.3, 128.2, 127.4, 125.8, 117.8, 109.3, 99.1, 56.8,
32.2, 31.3, 30.8, 30.2, 30.0, 30.0, 30.0, 30.0, 30.0, 29.7, 23.0, 14.4.
Compound 11 (65 mg, 0.087 mmol), K2CO3
(120 mg, 0.87 mmol) and p-toluensulfonate-tetraethylene glycol (65 mg, 0.181 mmol) were dissolved in DMF (15 mL) and the reaction mixture was heated to 85 °C for 16 h. After cooling down to rt, DMF was removed by rotary evaporation and the solid residue was dissolved in CH2Cl2 (40 mL), washed with distilled water (5 mL), washed
with brine (5 mL), dried over MgSO4, and concentrated in vacuo. Purification by
column chromatography (SiO2, CH2Cl2/MeOH 98:2) gave 12 (96 mg, 89%) as a
colorless oil; (Rf: 0.3 CH2Cl2/MeOH: 98:2); 1H NMR (400 MHz, CDCl3) δ 7.64 (s, 2H),
7.11 (s, 2H), 7.02 (d, J = 9.3 Hz, 2H), 6.88 (d, J = 9.3, 2H), 4.53 (d, J = 5.7 Hz, 2H), 4.39 (d, J = 5.7 Hz, 2H), 4.22 (t, J = 4.8 Hz, 4H), 3.89 (t, J = 4.9 Hz, 4H), 3.76 – 3.71 (m, 4H), 3.66 (tt, J = 7.9, 3.8 Hz, 16H), 3.58 – 3.50 (m, 4H), 3.36 (s, 6H), 2.96 – 2.89 (m, 4H), 2.86 – 2.73 (m, 8H), 1.83 – 1.69 (m, 4H), 1.44 – 1.23 (m, 36H), 0.87 (t, J = 6.6 Hz, 6H); 13C NMR (101 MHz, CDCl3) δ 156.3, 151.6, 137.1, 132.2, 128.5, 127.9, 127.7, 125.9, 118.7, 106.6, 99.1, 72.2, 71.1, 70.9, 70.9, 70.9, 70.8, 70.8, 70.0, 67.6, 59.3, 56.7, 32.2, 31.2, 30.8, 30.1, 30.0 (2C), 29.9, 29.9, 29.6, 23.0, 14.4; HRMS (ESI-ion trap) m/z: [M+H]+
Calcd for C66H107O14 1123.7661, found 1223.7645.
Compound 10 (300 mg, 0.345 mmol) was dissolved in dry THF (25 mL) and cooled to -78 °C under a nitrogen atmosphere. A solution of t-BuLi (1.2 mL, 1.55 mmol, 1.3 M in Pentane) was added dropwise and subsequently the reaction mixture was left stirring at -78 °C for 1 h, during which the color changed to dark brown. Dry DMF (120 µL, 1.55 mmol) was added dropwise causing the solution to turn colorless. The reaction mixture was slowly allowed to reach rt over the course of 16 h, quenched with sat. aq. NH4Cl,
extracted with EtOAc, dried over MgSO4 and concentrated in vacuo. Purification by
column chromatography (SiO2, Pentane/EtOAc 95:5 → Pentane/EtOAc 9:1) gave 13
(84.2 mg, 32%) as a colorless oil; (Rf: 0.4 Pent/ EtOAc 9:1); 1H NMR (400 MHz, CDCl3)
δ 10.13 (s, 2H), 8.35 (d, J = 1.6 Hz, 2H), 7.99 (s, 2H), 7.68 (dd, J = 8.8, 1.6 Hz, 2H), 7.21
(d, J = 8.8 Hz, 2H), 4.57 (d, J = 5.9 Hz, 2H), 4.50 (d, J = 5.8 Hz, 2H), 2.98 (m, 2H), 2.92 – 2.80 (m, 8H), 1.81 (m, 4H), 1.48 – 1.22 (m, 36H), 0.88 (t, J = 6.6 Hz, 6H); 13C NMR (101
MHz, CDCl3) δ 192.3, 156.5, 138.2, 136.2, 134.1, 133.6, 131.0, 130.4, 127.2, 125.7,
123.3, 99.3, 56.9, 32.2, 31.2, 30.6, 30.1, 30.0, 30.0, 30.0, 29.9 (2C), 29.7, 23.0, 14.4; HRMS (ESI-ion trap) m/z: [M+H]+ Calcd for C50H71O6 767.5245, found 767.5240.
89
5
Method A: Compound 13 (84 mg, 0.109 mmol) was dissolved in a 1:1 mixture of THF (2 mL) and MeOH (2 mL) and cooled to 0 °C. NaBH4 (12.4 mg, 0.327 mmol) was addedin small portions and the reaction mixture was left stirring at rt for 1 h. The reaction was quenched with sat. aq. NH4Cl,
the mixture was extracted with EtOAc dried over MgSO4 and
concentrated in vacuo. Purification by column chromatography (SiO2, Pentane/EtOAc 8:2 → Pentane/EtOAc 7:3) gave 14 (53.7 mg,
64%) as a colorless oil (Rf: 0.2 Pentane/EtOAc 8:2).
Method B: Compound 10 (115 mg, 0.13 mmol) was dissolved in dry THF (5 mL) and cooled to -78 °C. A solution of t-BuLi (0.34 mL, 0.58 mmol, 1.7 M in Pentane) was added dropwise and subsequently the reaction mixture was allowed to stir at -78 °C for 1 h, during which the color changed to dark brown. A suspension of paraformaldehyde (17 mg, 0.58 mmol) in dry THF (2 mL) was added dropwise to the reaction mixture. The mixture was slowly allowed to reach rt over the course of 16 h, quenched with sat. aq. NH4Cl, extracted with EtOAc, dried over MgSO4 and
concentrated in vacuo. Purification by column chromatography (SiO2, Pentane/EtOAc
8:2 → Pentane/EtOAc 7:3) gave 14 (85 mg, 85%) as a colorless oil (Rf: 0.2
Pentane/EtOAc 8:2); 1H NMR (400 MHz, CDCl3) δ 7.80 (s, 2H), 7.78 (s, 2H), 7.21 – 7.15
(m, 2H), 7.11 (d, J = 8.8 Hz, 2H), 4.80 (s, 2H), 4.55 (d, J = 5.7 Hz, 4H), 4.41 (d, J = 5.7 Hz, 4H), 2.97 (m, 2H), 2.90 – 2.77 (m, 8H), 1.83 – 1.64 (m, 4H), 1.38 (s, 36H), 0.88 (t, J = 6.6 Hz, 6H); 13C NMR (101 MHz, CDCl3) δ 153.6, 137.5, 137.0, 132.7, 131.1, 129.0, 126.8,
125.6, 125.4, 125.3, 99.2, 65.7, 56.8, 32.2, 31.2, 30.7, 30.1, 30.0, 30.0 (2C), 29.9 (2C), 29.7, 23.0, 14.4; HRMS (ESI-ion trap) m/z: [M+Na]+ Calcd for C50H74O6Na 793.5369,
found 793.5377.
NaH 60% (13 mg, 0.560 mmol) was suspended in 1 mL of dry THF under a nitrogen atmosphere. A solution of compound 14 (144 mg, 0.187 mmol) in dry THF (1 mL) was added to the NaH suspension and the mixture was heated at reflux for 1 h. A solution of p-toluensulfonate-tetraethylene glycol (169 mg, 0.467 mmol) in dry THF (2 mL) was added to the reaction mixture, which was left stirring while heated at reflux for 16 h. The mixture was allowed to reach rt, quenched with sat. aq. NH4Cl, extracted with
EtOAc dried over MgSO4 and concentrated in vacuo. Purification by column
chromatography (SiO2, CH2Cl2/MeOH 99:1 → CH2Cl2/MeOH 95:5) gave 15 (53.8 mg,
25%) as a colorless oil; (Rf: 0.8 CH2Cl2/MeOH 95:5); 1H NMR (400 MHz, CDCl3) δ 7.78 –
7.74 (m, 4H), 7.16 (d, J = 8.7 Hz, 2H), 7.08 (d, J = 8.7 Hz, 2H), 4.66 (s, 4H), 4.53 (d, J = 5.7 Hz, 2H), 4.39 (d, J = 5.6 Hz, 2H), 3.72 – 3.61 (m, 28H), 3.53 (dd, J = 5.8, 3.5 Hz, 4H), 3.36 (s, 6H), 2.96 (m, 2H), 2.88 (m, 8H), 1.84 – 1.73 (m, 4H), 1.46 – 1.23 (m, 36H), 0.88 (t, J = 6.8 Hz, 6H); 13C NMR (101 MHz, CDCl3) δ 153.5, 136.9, 134.9, 132.7, 131.0, 129.0,
90
5
126.6, 126.3, 125.9, 125.6, 99.2, 73.6, 72.2, 71.0, 70.9, 70.9, 70.9, 70.8 (2C), 69.9, 59.3, 56.8, 32.2, 31.2, 30.8, 30.2, 30.0, 30.0, 30.0, 30.0 (2C), 29.7, 23.0, 14.4; HRMS (ESI-ion trap) m/z: [M+NH4]+ Calcd for C68H114O14N1 1168.8234, found 1168.8250, [M+Na]+
Calcd for C68H110O14Na1 1173.7788, found 1173.7793.
Compound 15 (53 mg, 0.046 mmol) was dissolved in 1,4-dioxane (5 mL) under a nitrogen atmosphere. The reaction mixture was treated with aq. HCl (37%, 0.1 mL) and heated at reflux for 2 h. The mixture was then cooled down to rt, poured into distilled water (2 mL), extracted with CH2Cl2 (3x10 mL),
washed with brine, dried over MgSO4 and concentrated in vacuo. Purification by
column chromatography (SiO2, CH2Cl2/MeOH 95:5) gave 3 (42 mg, 86%) as a colorless
oil; (Rf: 0.6 CH2Cl2/MeOH 95:5); 1H NMR (400 MHz, CDCl3) δ 7.79 – 7.75 (m, 4H), 7.21 (d, J = 8.6, 2H), 7.02 (d, J = 8.6 Hz, 2H), 5.15 (s, 2H), 4.66 (s, 4H), 3.70 – 3.57 (m, 28H), 3.51 (dd, J = 5.8, 3.5 Hz, 4H), 3.34 (s, 6H), 2.85 (t, J = 7.7 Hz, 4H), 1.78 – 1.70 (m, 4H), 1.49 – 1.19 (m, 36H), 0.87 (t, J = 6.6 Hz, 6H); 13C NMR (101 MHz, CDCl3) δ 152.2, 134.0, 132.2, 131.9, 130.1, 129.5, 126.9, 126.8, 124.6, 111.1, 73.6, 72.2, 71.0, 70.9, 70.9, 70.9 (2C), 70.8, 69.7, 59.3, 32.2, 31.2, 30.0 (4C), 29.9 (2C), 29.9, 29.6, 23.0, 14.4; HRMS (ESI-ion trap) m/z: [M+NH4]+ Calcd for C64H106O12N1 1080.7738, found 1080.7710.
Compound 3 (35 mg, 0.032 mmol) was dissolved in pyridine (2 mL) and POCl3 (6
µL, 0.064 mmol) was added. The reaction mixture was left stirring at rt for 16 h, then diluted with CH2Cl2 (2 mL), washed
with aq. HCl (1M, 2 mL), dried over MgSO4 and concentrated in vacuo.
Purification by column chromatography (SiO2, CH2Cl2/MeOH 97:3) gave 16 (27 mg,
73%) as a colorless oil; (Rf: 0.5 CH2Cl2/MeOH 97:3); 1H NMR (400 MHz, CDCl3) δ 7.85
(m, 4H), 7.22 (m, 2H), 7.20 – 7.15 (m, 2H), 4.77 – 4.59 (m, 4H), 3.77 – 3.59 (m, 28H), 3.52 (dd, J = 5.9, 3.5 Hz, 4H), 3.35 (s, 6H), 3.22 – 3.04 (m, 2H), 2.97 – 2.79 (m, 2H), 1.88 – 1.69 (m, 4H), 1.47 – 1.18 (m, 36H), 0.92 – 0.80 (m, 6H); 31P NMR (162 MHz, CDCl3) δ
9.31; HRMS (ESI-ion trap) m/z: [M+NH4]+ Calcd for C64H104O13P1Cl1N1 1160.6928,
found 1160.6983.
Compound 16 (27 mg, 0.024 mmol) was dissolved in THF (2 mL) and treated with aq. HCl (1M, 2 mL). The reaction mixture was left stirring at rt for 2 h, then extracted with CH2Cl2 ), dried over MgSO4
and concentrated in vacuo. Purification by column chromatography (SiO2,
91
5
CH2Cl2/MeOH 95:5) gave 17 (25 mg, 96%) as a colorless oil; (Rf: 0.2 CH2Cl2/MeOH
97:3); 1H NMR (400 MHz, CDCl3) δ 7.80 (s, 2H), δ 7.78 (s, 2H), 7.22 (d, J = 8.8 Hz, 2H),
7.17 (d, J = 8.8, 2H), 4.74 – 4.61 (m, 4H), 3.71 – 3.57 (m, 28H), 3.51 (dd, J = 5.8, 3.5 Hz, 4H), 3.34 (s, 6H), 3.10 (m, 2H), 2.84 (m, 2H), 1.77 (m, 4H), 1.39 (m, 4H), 1.34 – 1.15 (m, 32H), 0.85 (t, J = 6.9 Hz, 6H); 31P NMR (162 MHz, CDCl3) δ 4.37; HRMS (ESI-ion trap)
m/z: [M+H]+ Calcd for C64H102O14P1 1125.7002, found 1125.7005, [M+Na]+ Calcd for
C64H101O14P1Na1 1147.6821, found 1147.6817.
5.1.9 References
[1] J. M. Brunel, Chem. Rev. 2007, 107, 1–45.
[2] R. Maji, S. C. Mallojjala, S. E. Wheeler, Chem. Soc. Rev. 2018, 47, 1142–1158. [3] T. Akiyama, K. Mori, Chem. Rev. 2015, 115, 9277–9306.
[4] T. Akiyama, Chem. Rev. 2007, 107, 5744–5758.
[5] T. Akiyama, J. Itoh, K. Yokota, K. Fuchibe, Angew. Chem. Int. Ed. 2004, 43, 1566–1568. [6] J. H. Kim, I. Čoric̈, S. Vellalath, B. List, Angew. Chem. Int. Ed. 2013, 52, 4474–4477. [7] M. Terada, H. Tanaka, K. Sorimachi, J. Am. Chem. Soc. 2009, 131, 3430–3431. [8] K. Mori, K. Ehara, K. Kurihara, T. Akiyama, J. Am. Chem. Soc. 2011, 133, 6166–6169. [9] D. Uraguchi, K. Sorimachi, M. Terada, J. Am. Chem. Soc. 2004, 126, 11804–11805. [10] J. Zhang, S. X. Lin, D. J. Cheng, X. Y. Liu, B. Tan, J. Am. Chem. Soc. 2015, 137, 14039–
14042.
[11] D. Parmar, E. Sugiono, S. Raja, M. Rueping, Chem. Rev. 2014, 114, 9047–9153.
[12] M. Fleischmann, D. Drettwan, E. Sugiono, M. Rueping, R. M. Gschwind, Angew. Chem. Int.
Ed. 2011, 50, 6364–6369.
[13] B. Schuur, B. J. V. Verkuijl, J. Bokhove, A. J. Minnaard, J. G. De Vries, H. J. Heeres, B. L. Feringa, Tetrahedron 2011, 67, 462–470.
[14] E. B. Pinxterhuis, J. B. Gualtierotti, H. J. Heeres, J. G. De Vries, B. L. Feringa, Chem. Sci.
2017, 8, 6409–6418.
[15] E. B. Pinxterhuis, J. B. Gualtierotti, S. J. Wezenberg, J. G. de Vries, B. L. Feringa,
ChemSusChem 2018, 11, 178–184.
[16] R. Mitra, M. Thiele, F. Octa-Smolin, M. C. Letzel, J. Niemeyer, Chem. Commun. 2016, 52, 5977–5980.
[17] R. Mitra, H. Zhu, S. Grimme, J. Niemeyer, Angew. Chem. Int. Ed. 2017, 56, 11456–11459. [18] A. C. Coleman, J. M. Beierle, M. C. A. Stuart, B. Maciá, G. Caroli, J. T. Mika, D. J. van Dijken,
J. Chen, W. R. Browne, B. L. Feringa, Nat. Nanotechnol. 2011, 6, 547–552.
[19] D. J. van Dijken, P. Štacko, M. C. A. Stuart, W. R. Browne, B. L. Feringa, Chem. Sci. 2017, 8, 1783–1789.
[20] D. J. van Dijken, J. Chen, M. C. A. Stuart, L. Hou, B. L. Feringa, J. Am. Chem. Soc. 2016, 138, 660–669.
[21] D. Ishikawa, T. Mori, Y. Yonamine, W. Nakanishi, D. L. Cheung, J. P. Hill, K. Ariga, Angew.
Chem. Int. Ed. 2015, 54, 8988–8991.
[22] C. Wendeln, S. Rinnen, C. Schulz, T. Kaufmann, H. F. Arlinghaus, B. J. Ravoo, Chem. Eur. J.
92
5
93
5
Section II – Salen-Based Amphiphiles
for Catalysis
5.2.1 Introduction
Salen metal complexes represent one of the most exploited classes of ligands in catalysis. Their synthetic versatility was shown by Jacobsen and co-workers who reported the stereoselective epoxidation of alkenes catalysed by Mn(III) salen complex.[1] In one of the first studies, systematic testing led to the discovery of the
chiral scaffold derived from trans-1,2-cyclohexyldiamine as the most effective ligand, although the use of 1,2-diphenylethanediamine also provided good results. The significance of the design lies in the simplicity to access both enantiomers of the ligand in a straightforward two-step synthesis from commercially available starting materials, which made the system highly popular in the catalysis community.[2–6]
Moreover, the simplicity of their synthesis and their modularity have made them very attractive systems for supramolecular chemistry, as described in Chapter 3.[7]
This initial design inspired the Jacobsen group and many others towards the synthesis of several analogues, by changing metal centers and introducing various substituents at the scaffold.[8–11] The versatility in metal complexation made salen
ligands very popular. Mainly focussing on the first row of transition metals, many systems have been developed to catalyse reactions such as hydrolytic kinetic resolution of epoxides,[9] epoxide ring-opening,[8] nucleophilic additions,[12]
stereoselective oxidations[13] and many more.[14–16]
Some of these catalytic systems have been reported to be effective in double activation of substrates.[11] For example, Co-salen complexes, synthesized as oligomers
in confined systems, proved to be more active than the monomeric species revealing an activation pathway which involved more than one metal center in the stereoselective epoxide ring-opening reaction.
In our quest to perform stereoselective catalysis in self-assembled systems in water, we directed our attention towards metal-salen complexes. As described in Chapter 3, we managed to successfully synthesize the complexes of Cu, Ni, Co, Fe and Mn of a novel amphiphilic salen ligand L1 (Figure 5.2.1).
94
5
We reported the detailed study of the self-assembled nanostructures provided by this system in water, resulting in stable aggregates which have great potential for catalytic application in aqueous environment. In the design of the amphiphiles, C12
chains were positioned in close proximity to the chelating site where the metal center is complexed. Our intention was to have the catalytic site pointing towards the lipophilic phase generated in the self-assembly process, where organic substrates can be encapsulated. Here we describe our preliminary studies towards application of the novel amphiphiles in various asymmetric transformations.
5.2.2 Henry Reaction
Salen complexes of Ni and Cu have been reported for the stereoselective Henry reaction between nitroalkanes and aldehydes, and the positive effect of their inclusion in a confined reaction environment, such as within zeolites, has been demonstrated.[17–19] The catalysis reported was performed in both protic and aprotic
organic solvents, but not in water. Therefore, it would be interesting to investigate the effectiveness of this reaction in aqueous environment, promoted by the use of amphiphilic catalysts.
The study on the aggregation of amphiphilic salen complexes L1-Cu and L1-Ni described in Chapter 3 revealed the formation of inverted structures in water, described as sponges. The tightly packed network of inverted structures has been reported to effectively include organic substrates,[20,21] which is a necessary
requirement to perform catalysis. Our attempts to perform catalysis in these systems are described in Table 5.2.1.
Entry Complex Loading Temperature Additive Conversion (3)a e.e. (%)
1 L1-Ni 1% rt - 62% rac.
2 L1-Ni 5% rt - 64% rac.
3 L1-Ni 5% rt TEA 96% rac.
4 L1-Cu 1% rt - 65% rac.
5 L1-Cu 5% rt - 68% rac.
6 L1-Cu 5% rt TEA 96% rac.
7b L1-Ni 5% rt - 67% rac. 8b L1-Cu 5% rt - 68% rac. 9 L1-Ni 5% 4 °C - 50% rac. 10 L1-Cu 5% 4 °C - 52% rac. 11c - - rt - 62% - 12c - - rt TEA 95% - 13 - - 4 °C - 48% - 14 - - 4 °C TEA 90% -
Table 5.2.1: Attempts to perform stereoselective Henry reaction with L1-Cu and L1-Ni in H2O. Reaction
conditions: 1 equiv. of 4-Nitrobenzaldehyde was added to a solution containing self-assembled sponge aggregates followed by 10 equiv. of nitromethane, and the reaction mixture was left stirring for 16 h; a)
95
5
Determined by 1H NMR; b) 4-nitrobenzaldehyde was added before the self-assembly process; c) Thereaction was also performed using new glass vials and stirring bars, showing comparable results.
The reaction was initially attempted using L1-Ni, which did not show leaching of the catalyst in EDX measurements (Chapter 3, Figure 3.4), and therefore was considered a more stable complex to use.
Using low catalyst loading, the product 3 of the Henry reaction between nitromethane and 4-nitrobenzaldehyde with L1-Ni was obtained with 62% conversion (Table 5.2.1, entry 1). Increasing the catalyst loading did not prove beneficial to the conversion (entry 2), but the addition of an organic base such as triethylamine (TEA) succeeded in increasing the conversion to 96% (entry 3). Unfortunately, in all cases the product was obtained as a racemic mixture. Performing the same investigation with L1-Cu provided similar results in terms of conversion (entries 4-6), and also in these cases compound 3 was obtained as a racemic mixture.
It is worth mentioning that in all cases, the turbid aqueous sample containing the self-assembled metallo-amphiphile underwent phase separation after addition of the reactants, where the liquid reactants formed an “organic phase” on top of the water layer. The so-formed organic phase presented intense colour (red for L1-Ni and brown for L1-Cu) suggesting that the metallo-amphiphile was extracted from the water, and therefore de-aggregated. In fact, the water phase lost the typical colour of the complexes and turned less turbid than in presence of the self-assembled sponges.
Since cubic phases are not dynamic aggregates in solution, it was anticipated that the addition of reagents after the self-assembly process did not result in the inclusion of substrate in the soft material. Assuming the problem would be due to the instability of the aggregate upon addition of other reagents, the reaction was tested by adding the electrophile to L1-Cu and L1-Ni before the self-assembly process took place (entries 7-8). Both tests provided comparable results when the addition of the substrate was done before the self-assembly process.
Suspecting that the addition of the reactants would influence the stability of the aggregates at room temperature, the reaction was performed at 4 °C (entries 9-10). Unfortunately, phase separation occurred in this case as well, obtaining lower conversion to the product than at room temperature but not achieving any stereoselectivity.
Observing the formation of racemic product in all cases, a reaction in pure water in absence of the metallo-amphiphile was performed as a control. The product was obtained with 62% conversion by using pure water at room temperature and 48% at 4 °C (entries 11 and 13), and by addition of TEA the conversion was increased to 95% at room temperature and 90 at 4 °C (entries 12 and 14). To exclude the influence on the catalysis of trace transition metal particles in the stirring bar, or eventual contamination from previous experiments, the reaction was performed in newly purchased vials and with new stirring bars. The results were reproducible, revealing that the Henry reaction proceeded on water[22] smoothly without the need of the
metallo-amphiphile at the tested conditions.
The reaction seemed to occur in the absence of the catalyst on water, and furthermore phase separation due to the addition of reactants hampered the catalytic test in the self-assembled sponges. Undoubtedly, inverted structures such as sponges show little permeability of substrates and are therefore less explored in catalysis, with
96
5
the exception of a couple of examples reported in recent years mentioned in Chapter 1.[23,24] Because of these reasons, we decided to move onto other test reactions using
the related salen-based amphiphiles which aggregated in vesicular and micellar assemblies (see Chapter 3).
5.2.3 Ring-Opening of Epoxides
Having the metallo-amphiphile L1-Co described in Chapter 3 in hand, which self-assembled into vesicles, we thought to investigate its catalytic application in epoxide-opening reactions.
Salen-cobalt complexes are very well-explored catalysts, which became increasingly popular with the publication of the pioneering example of the efficient Hydrolytic Kinetic Resolution (HKR) of epoxides with water as a nucleophile.[9] This
reaction has been further performed in water as the only solvent using an amphiphilic polymer which featured salen pendants functionalized with Co(III) metal ions.[25] The
self-assembly of the reported amphiphilic polymer revealed the formation of spherical aggregates via TEM, confirmed by DLS measurements, which proved to be an efficient catalytic system achieving high e.e. values. Expecting that our system could be successfully implemented for this transformation, we performed the tests summarized in Table 5.2.2.
Entry Time % THFa L1-Co Conversion (5)b e.e. (%)
1 1 h 5 0.5% <5% rac. 2 2 h 5 0.5% 7% rac. 3 20 h 5 0.5% 15% rac. 4 20 h 20 0.5% 15% rac. 5 20 h 5 1% 17% rac. 6 20 h 5 5% 16% rac. 7c 20 h 0 - 8% rac.
Table 5.2.2: Attempts to perform HKR with L1-Co in H2O. Reaction conditions: 1 equiv. of styrene oxide
was added to a solution containing self-assembled vesicle aggregates, and the reaction mixture was left stirring for a chosen amount of time; a) THF added in the self-assembly process; b) Determined by 1H NMR;
c) The reaction was also performed using new glass vials and stirring bars, showing comparable results.
The reaction was initially investigated utilizing the vesicles obtained by the same preparation method as for the Cryo-TEM samples (described in Chapter 3). A 0.5 mol% catalyst loading was chosen, as described in the above-reported methods.[9,25]
The reaction did not achieve good conversion to product after 1 h (Table 5.2.2, entry 1), therefore a longer reaction time was tested. After 2 or 20 h the conversion did not increase significantly, providing only 15% conversion to compound 5 (entries 2-3). Increasing the THF content in the system from 5 to 20% did not improve the conversion (entry 4), and in all cases the products were obtained as a racemic mixture (compounds 5 and 4). Moreover, in this case phase separation also occurred and the
97
5
organic phase generated by the liquid substrate assumed a brown color, characteristic of the cobalt complex L1-Co. Even in this case, the assembly de-aggregated in the presence of the substrate (vide supra). This resulted in difficulty in operation since water acts as a reagent in this system and phase separation perhaps hampered the reaction.
It was considered that one of the problems could have been the low amphiphile catalyst loading and therefore we decided to increase the amount of L1-Co in the test reaction. Increasing the loading to 1 and 5 mol% (entries 5-6) did not improve either the conversion nor the stereoselectivity. The control reaction in this case was also performed in pure water, with clean glassware and a new stirring bar, and it provided a slightly lower conversion to product than in the presence of the amphiphile (entry 7). Apparently, the presence of the amphiphile somehow slightly increased the conversion, but no transfer of stereochemical information to the product was observed.
The conversion was very low, probably due to the design of the amphiphile itself. The complex L1-Co was synthesized featuring alkyl chains pointing towards the metal center in order to favor reactivity with lipophilic and organic substrates. The epoxide may be included in the lipophilic phase which means that it would not come in contact with the aqueous phase, which would hamper the reaction since water acts as a reagent.
For this reason, we decided to test catalysis for the ring-opening of epoxides with organic nucleophiles.[8,26] The desymmetrization of meso-epoxides with anilines has
been reported to be catalyzed by Co and Fe salen complexes, which provided us with a further possibility to find application for metallo-amphiphiles L1-Co and L1-Fe. The reaction tests are summarized in Table 5.2.3.
Entry Complex Loading Temperature Conversion (8)a e.e. (%)
1 L1-Co 1% rt Full rac.
2 L1-Fe 1% rt Full rac.
3b L1-Co 1% rt Full rac.
4b L1-Fe 1% rt Full rac.
5 L1-Co 1% 4 °C Full rac.
6 L1-Fe 1% 4 °C Full rac.
7c - - rt Full rac.
8c - - 4 °C Full rac.
9d - - rt - -
Table 5.2.3: Attempts to perform the desymmetrization of cyclohexene oxide with L1-Co and L1-Fe.
Reaction conditions: 1 equiv. of cyclohexene oxide was added to a solution containing self-assembled vesicle aggregates followed by 1 equiv. of aniline, and the reaction mixture was left stirring for 16 h; a) Determined by 1H NMR; b) Addition of the reagents before the self-assembly process; c) The reaction was
also performed using new glass vials and stirring bars, showing comparable results; d) Reaction performed neat.
98
5
In presence of both L1-Co and L1-Fe full conversion to product 8 was observed, using 1 mol% of complex at room temperature and adding the reagents to a vesicle solution of L1-Co or L1-Fe (Table 5.2.3, entries 1-2). In all cases the product obtained was racemic, and phase separation occurred after the addition of the reactants (vide
supra). The reaction was tested by adding the reactants before performing
self-assembly (entries 3-4). Although the outcome did not show any difference, slower phase separation was observed when the reagents were added prior to self-assembly. The reaction was also tested at lower temperatures (entries 5-6), still providing full conversion and racemic product. The reaction always proceeded smoothly and, since product 8 was obtained as a racemic mixture, control experiments in the absence of the complexes were performed (entries 7-8).
Although the ring opening of epoxides by anilines is not a fast reaction in organic solvents, in absence of the metallo-amphiphile in water it also proceeded to full conversion (entries 7-8). Testing with new glassware and stirring bar provided the same result, showing that the reaction was not catalysed by trace metal impurities. Performing the control experiment in the absence of solvent (entry 9) did not provide any conversion to the product, suggesting that the reaction proceeded quickly on water. The presence of water apparently favours the ring opening of the epoxide by aniline, which may be not only because of increased local concentration but also due to possible H-bonding of the water molecules at the interface between the phase generated by the reagents and the aqueous environment.
5.2.4 Alkene Epoxidation
The epoxidation of alkenes denotes one of the first reports of catalysis with complexes deriving from salen-type ligands.[1] After having successfully synthesized
the Mn(III)Cl complex from the amphiphilic ligand L1, which self-assembled into spherical micelles, we thought to use this example in reaction tests. In the original studies, analogous ligands based on 1,2-diphenylethanediamine have been reported for the same reaction, which encouraged us regarding the outcome of the catalysis. Preliminary tests are shown in Table 5.2.4.
Entry Substrate L1-Mn Loading Oxidanta Oxidant (equiv.) Conversion (10) b
1 9a 10% H2O2 (30%) 5 - 2 9a 10% H2O2 (10%)c 5 - 3 9a 5% H2O2 (30%) 5 - 4 9a 10% H2O2 (30%) 3 - 5 9b 10% H2O2 (30%) 5 - 6 9a 10% NaOCl (4%)d 5 -
Table 5.2.4: Attempts to perform stereoselective epoxidation with L1-Mn; a) Percentages expressed in
99
5
All the reactions were tested generating a micellar suspension of L1-Mn in doubly distilled H2O at room temperature. The benchmark substrates chosen for the
test are commonly used tri-substituted alkenes for this transformation. Initially, the epoxidation of 1,2-dihydronaphthalene 9a was tested in the presence of 10 mol% L1-Mn with 5 equivalents of H2O2 (30%) (Table 5.2.4, entry 1). Addition of the oxidant
caused the micellar solution to change color from purple to dark brown within seconds, followed by phase separation as in the previously discussed tests with the other salen-amphiphiles. This observation suggested that oxidation at the metal center might have happened, but unfortunately the formation of the epoxide product was not detected.
Lowering the concentration of H2O2 oxidant to 10% and adding it dropwise to the
reaction mixture did not improve the conversion (entry 2). Further testing with a lower amphiphile loading (5 mol%) or less H2O2 (30%) (3 equivalents) did not prove
successful either (entries 3 and 4, respectively). Indene 9b was also tested as a substrate with of 10 mol% L1-Mn and 5 equivalents of H2O2 (30%), but the reaction
did not yield any product.
It was reasoned that the problem in this system might have been due to the oxidant. Therefore, the epoxidation of 9b was tested with NaOCl (5 equivalents) using 10 mol% of L1-Mn, but it did not result in any product formation. The outcome in all the reactions tested was similar, obtaining two separate phases shortly after adding the reagents to the micellar solution of L1-Mn. No epoxide product formation was detected in any of the reaction mixtures after stirring overnight, as the starting alkene remained unreacted.
5.2.5 Conclusions
Attempts to perform catalysis with the salen-based amphiphiles reported in Chapter 3 were performed. Unfortunately, in the cases of Cu, Ni, Co or L1-Fe the reactions proceeded fast on water without the need of the metallo-amphiphile, thus resulting in the formation of racemic products. The catalytic epoxidation of alkenes was tested with L1-Mn, but the formation of the product was not detected. In the latter case, addition of the oxidant caused the reaction mixture to change color suggesting that oxidation at the metal center occurred, but the starting material was recovered unreacted. In all the above-mentioned cases, phase separation was observed after the addition of the reactants, revealing that the self-assembled structures were not stable under the tested conditions. These experiments cannot showcase the application of these promising novel amphiphiles in surfactant catalysis and the reactions need further investigation. A possible problem could be represented by the tight packing of the metallo-amphiphiles, which might be overcome by designing amphiphilic systems featuring a spacer which allows for better permeability of the substrates to the catalytic metal center. Alternatively, co-assembly with other surfactants could provide an opportunity for a successful catalytic system in water, although it would require the use of an overall higher amount of amphiphile.
100
5
5.2.6 Contributions
The projects were designed jointly by F. Tosi and prof. dr. B. L. Feringa. The catalytic testing of the Henry reaction and epoxide ring-opening reactions was carried out by F. Tosi. The manganese-catalyzed alkene epoxidation was carried out by dr. R. Dorel.
5.2.7 Experimental Procedures
Analytical data of compounds 3,[27] 4,[28] 5[29] and 8[30] were found in accordance
with those reported.
Testing of the Henry reaction. Amphiphile L1-Ni (1.17 mg, 0.001 mmol) or L1-Cu (1.23 mg, 0.001 mmol) was dissolved in THF (50 μL), added to doubly distilled H2O (1 mL) and shock
frozen in liquid nitrogen and sonicated for three consecutive cycles. To the amphiphile solution 4-nitrobenzaldehyde 2 (12 mg, 0.082 mmol) was added followed by the addition of nitromethane (44 μL, 49 mg, 0.82 mmol) and the reaction mixture was left stirring for 16 h. The mixture was extracted with EtOAc (2 mL), dried over MgSO4, filtered over neutral Al2O3 and
concentrated in vacuo. The conversion was determined by 1H NMR analysis.
Testing of the Henry reaction, with substrate addition before self-assembly. Amphiphile L1-Ni (1.17 mg, 0.001 mmol) or L1-Cu (1.23 mg, 0.001 mmol) was dissolved in THF (50 μL) and 4-nitrobenzaldehyde 2 (12 mg, 0.082 mmol) was added. To the mixture, doubly distilled H2O (1 mL) was added and the sample was shock
frozen in liquid nitrogen and sonicated for three consecutive cycles. Nitromethane (44 μL, 49 mg, 0.82 mmol) was added to the reaction mixture which was left stirring for 16 h. The mixture was extracted with EtOAc (2 mL), dried over MgSO4, filtered over
neutral Al2O3 and concentrated in vacuo. The conversion was determined by 1H NMR
analysis. Compound 3: 1H NMR (400 MHz, CDCl3) δ 8.26 (d, J = 8.5 Hz, 2H), 7.62 (d, J =
8.5 Hz, 2H), 5.66 – 5.55 (m, 1H), 4.58 (m, 2H), 3.30 (br, 1H). The spectral data was found in agreement with the literature.[27]
Testing of the Hydrolytic Kinetic Resolution. Amphiphile L1-Co (1.28 mg, 0.001 mmol) was dissolved in THF (50 μL), added to doubly distilled H2O (1 mL) and shock frozen in liquid nitrogen and
sonicated for three consecutive cycles. Styrene oxide 4 (11 μL, 12 mg, 0.10 mmol) was added to the amphiphile solution and the reaction mixture was left stirring for a chosen amount of time (1, 2 or 20 h) at rt. The mixture was extracted with EtOAc (2 mL), the organic phase was dried over MgSO4,
filtered over neutral Al2O3 and concentrated in vacuo. The conversion was determined
101
5
4.84 (m, 1H), 3.77 (m, 1H), 3.66 (m, 1H), 2.47 (br, 1H), 2.01 (br, 1H). Compound 4: 1H
NMR (400 MHz, CDCl3) δ 7.39 – 7.26 (m, 5H), 3.87 (dd, J = 4.1, 2.6 Hz, 1H), 3.15 (dd, J =
5.5, 4.1 Hz, 1H), 2.81 (dd, J = 5.5, 2.6 Hz, 1H). The spectral data was found in agreement
with the literature.[28,29]
Testing of the epoxide desymmetrization. Amphiphile L1-Co (1.28 mg, 0.001 mmol) or L1-Fe (1.25 mg, 0.001 mmol) was dissolved in THF (50 μL), added to doubly distilled H2O (1 mL) and shock frozen in liquid
nitrogen and sonicated for three consecutive cycles. To the amphiphile solution cyclohexene oxide 6 (10 μL, 9.8 mg, 0.10 mmol) was added followed by the addition of aniline 7 (6.0 μL, 9.3 mg, 0.10 mmol) and the reaction mixture was left stirring for 16 h. The mixture was extracted with EtOAc (2 mL), the organic phase was dried over MgSO4, filtered over neutral Al2O3 and
concentrated in vacuo. The conversion was determined by 1H NMR analysis.
Testing of the epoxide desymmetrization, with substrate addition before self-assembly. Amphiphile L1-Co (1.28 mg, 0.001 mmol) or L1-Fe (1.25 mg, 0.001 mmol) was dissolved in THF (50 μL), followed by the addition of cyclohexene oxide 6 (10 μL, 9.8 mg, 0.10 mmol) and aniline 7 (6.0 μL, 9.3 mg, 0.10 mmol). To the mixture, doubly distilled H2O (1 mL) was added and the sample was shock frozen in liquid nitrogen
and sonicated for three consecutive cycles. The reaction mixture was left stirring for 16 h, then extracted with EtOAc (2 mL), the organic phase was dried over MgSO4,
filtered over neutral Al2O3 and concentrated in vacuo. The conversion was determined
by 1H NMR analysis. Compound 8: 1H NMR (400 MHz, CDCl3) δ 7.22 – 7.11 (m, 2H),
6.78 – 6.64 (m, 3H), 3.36 (td, J = 9.8, 4.3 Hz, 1H), 3.15 (ddd, J = 10.8, 9.2, 4.0 Hz, 1H), 2.75 (br, 1H), 2.18 – 2.06 (m, 2H), 1.84 – 1.68 (m, 2H), 1.45 – 1.23 (m, 3H), 1.14 – 0.99 (m, 1H). The spectral data was found in agreement with the literature.[30]
Testing of the stereoselective epoxidation. Amphiphile L1-Mn (1.25 mg, 0.001 mmol) was dissolved in THF (50 μL), added to doubly distilled H2O (1 mL) and shock
frozen in liquid nitrogen and sonicated for three consecutive cycles. To the amphiphile solution 1,2-dihydronaphthalene 9a (6.5 mg, 0.050 mmol) or indene (, 5.8 mg, 0.050 mmol) were added followed by the addition of the oxidant (5 or 10 equiv.) and the reaction mixture was left stirring for 16 h. The mixture was extracted with EtOAc (2 mL), the organic phase was dried over MgSO4 and concentrated in vacuo. No
conversion to product was observed by 1H NMR analysis.
5.2.8 References
[1] E. N. Jacobsen, W. Zhang, A. R. Muci, J. R. Ecker, L. Deng, J. Am. Chem. Soc. 1991, 113, 7063–7064.
[2] S. Shaw, J. D. White, Chem. Rev. 2019, 119, 9381–9426. [3] L. Canali, D. C. Sherrington, Chem. Soc. Rev. 1999, 28, 85–93.
[4] C. Freire, M. Nunes, C. Pereira, D. M. Fernandes, A. F. Peixoto, M. Rocha, Coord. Chem.
Rev. 2019, 394, 104–134.
[5] A. K. Shiryaev, Curr. Org. Chem. 2012, 16, 1788–1807.
[6] M. Abd El Sater, N. Jaber, E. Schulz, ChemCatChem 2019, 11, 3662–3687.
102
5
10.1002/anie.201908010.
[8] E. N. Jacobsen, Acc. Chem. Res. 2000, 33, 421–431.
[9] M. Tokunaga, J. F. Larrow, F. Kakiuchi, E. N. Jacobsen, Science 1997, 277, 936–938. [10] S. Jónsson, F. G. J. Odille, P. O. Norrby, K. Wärnmark, Org. Biomol. Chem. 2006, 4, 1927–
1948.
[11] J. M. Ready, E. N. Jacobsen, J. Am. Chem. Soc. 2001, 123, 2687–2688. [12] J. D. White, S. Shaw, Chem. Sci. 2014, 5, 2200–2204.
[13] Z. Yang, C. Zhu, Z. Li, Y. Liu, G. Liu, Y. Cui, Chem. Commun. 2014, 50, 8775–8778. [14] C. Baleizão, H. Garcia, Chem. Rev. 2006, 106, 3987–4043.
[15] P. G. Cozzi, Chem. Soc. Rev. 2004, 410–421. [16] T. Katsuki, Adv. Synth. Catal. 2002, 344, 131–147.
[17] M. Sharma, B. Das, G. V. Karunakar, L. Satyanarayana, K. K. Bania, J. Phys. Chem. C 2016,
120, 13563–13573.
[18] K. K. Bania, G. V. Karunakar, B. Sarma, R. C. Deka, Chempluschem 2014, 79, 427–438. [19] K. K. Bania, G. V. Karunakar, K. Goutham, R. C. Deka, Inorg. Chem. 2013, 52, 8017–8029. [20] T. Zhou, J. J. Vallooran, S. Assenza, A. Szekrenyi, P. Clapés, R. Mezzenga, ACS Catal. 2018,
8, 5810–5815.
[21] M. Duss, L. Salvati Manni, L. Moser, S. Handschin, R. Mezzenga, H. J. Jessen, E. M. Landau,
ACS Appl. Mater. Interfaces 2018, 10, 5114–5124.
[22] R. Ballini, G. Bosica, J. Org. Chem. 1997, 62, 425–427.
[23] T. Zhou, J. J. Vallooran, S. Assenza, A. Szekrenyi, P. Clapés, R. Mezzenga, ACS Catal. 2018,
8, 5810–5815.
[24] M. Duss, L. Salvati Manni, L. Moser, S. Handschin, R. Mezzenga, H. J. Jessen, E. M. Landau,
ACS Appl. Mater. Interfaces 2018, 10, 5114–5124.
[25] B. M. Rossbach, K. Leopold, R. Weberskirch, Angew. Chem. Int. Ed. 2006, 45, 1309–1312. [26] S. Roy, P. Bhanja, S. S. Islam, A. Bhaumik, S. M. Islam, Chem. Commun. 2016, 52, 1871–
1874.
[27] A. Ebru Aydin, S. Yuksekdanaci, Tetrahedron: Asymmetry 2013, 24, 14–22. [28] W. Dai, J. Li, G. Li, H. Yang, L. Wang, S. Gao, Org. Lett. 2013, 15, 4138–4141.
[29] K. Wu, L. Chen, H. Fan, Z. Zhao, H. Wang, D. Wei, Tetrahedron Lett. 2016, 57, 899–904. [30] K. Kamińska, E. Wojaczyńska, J. Skarżewski, A. Kochel, J. Wojaczyński, Tetrahedron:
