Solid-phase lanthanum catalysis of monoacylation of diols in water by acyl phosphate monoesters

Article abstract of DOI:10.1139/cjc-2014-0406

Lanthanide ions are readily bound to ion-exchange resins. The lanthanide-containing resins serve as immobilized catalysts for the biomimetic monoacylation of diols in water using acyl phosphate monoesters as acylation agents. This method provides an efficient route for recovering the catalyst in the process of modifying RNA derivatives and carbohydrates.

Full text of DOI:10.1139/cjc-2014-0406

Page 1 of 15
1
Solid phase lanthanum catalysis of monoacylation of diols in water by acyl phosphate
monoesters
Raj S. Dhiman, Aizhou Wang and Ronald Kluger*
Davenport Chemistry Research Laboratories, Department of Chemistry, University of Toronto,
Toronto, ON M5S 3H6
*Corresponding author: Ronald Kluger, email: r.kluger@utoronto.ca, tel: 416ꢀ978ꢀ3582
Dedicated to Professor Stan Brown, a pioneer of biomimetic metal ion chemistry

Page 2 of 15
2
Abstract
Lanthanide ions are readily bound to ionꢀexchange resins. The lanthanideꢀcontaining resins serve
as immobilized catalysts for the biomimetic monoacylation of diols in water using acyl
phosphate monoesters as acylation agents. This method provides an efficient route for recovering
the catalyst in the process of modifying RNA derivatives and carbohydrates.
Keywords: Lanthanide catalysis, solid phase, biomimetic aminoacylation
Introduction
Lanthanum has been developed as a catalyst for biomimetic monoacylation of diols by acyl
phosphate monoesters, presumably occurring via coꢀchelation followed by lanthanum promoted
acylation. The general approach mimics the biological system that utilizes enzymically produced
aminoacyl adenylates for aminoacylation of the 3’ꢀterminus of tRNA. The method has been
shown to work successfully on nucleosides, nucleotides, RNA1 and monosaccharides. 2, 3 This
approach is distinct from other diol acylation methods, as esters will be produced through bisꢀ
bidentate chelates of lanthanum as shown in Scheme 1. The method has also been extended by
Duffy and Dougherty to aminoacylation of fullꢀlength suppressor tRNAs, using the resulting
aminoacylꢀtRNAs in protein synthesis for their studies of ion channels. 4
The waterꢀsolubility of bioꢀmolecules and the unique regiospecificity of lanthanumꢀpromoted
acylation have the potential for wide use in the selective acylation of biomolecules. However, the
process has a low efficiency in terms of lanthanum. Duffy and Dougherty report that a very large
excess of the lanthanide is needed to produce a detectable yield of product, making isolation of a
4
stable product challenging. In any case, even with less lanthanum, the metal ion must be
completely removed from the product for further work as lanthanides promote the cleavage of
RNA. As lanthanides are waterꢀsoluble, subsequent separation from the reaction mixture is a
complex process requiring quenching of the lanthanide with DTPA followed by sizeꢀexclusion
chromatography. Moreover, lanthanide salts are widely used in general as Lewis acids in water
and can be recycled if recovered. 5 In the existing biomimetic acylation process, the lanthanide is
difficult to separate from the product and is not readily recovered for reꢀuse.

Page 3 of 15
3
Scheme 1. Lanthanumꢀmediated acylation of a cis-diol: bisꢀbidentate coordination of BMP
(benzoyl methyl phosphate) and a cis-diol about a lanthanide center followed by acyl transfer
completes the reaction.
Immobilization of the lanthanide ion onto a polymer support is potentially an effective
solution for convenient separation and reꢀuse of the catalyst. Due to the established affinity of
polyols for lanthanides, lanthanideꢀmodified ionꢀexchange resins have been employed by Angyal
6
for carbohydrate separations. Angyal’s procedure uses water as a solvent and depends on
coordination of the hydroxyl groups of the sugar to the immobilized metal ion, which is similar
to the basis of the affinity we would use for an immobilized catalyst for reactions of diols in
water. In principle, the addition of an acyl phosphate monoester would then convert a solid
support designed for separations into a recoverable reagent for biomimetic acylation. This
approach should be practical as ionꢀexchange resins are readily available and are relatively
inexpensive. The use of the immobilized catalyst would simplify the separation of the lanthanide
and avoid the need for a chromatographic isolations. This approach also has the added benefit of
recycling of the lanthanide, which reduces waste and is desirable in terms of environmental
considerations. 7
A similar approach to recovery and recycling of lanthanides for aqueous aldol reactions has
been reported by Kobayashi. 8 However, that approach requires synthesis of the polymer support.

Page 4 of 15
4
A related study by Yu et. al. describes the preparation and evaluation of lanthanides bound to ion
exchange resins as catalysts for several reactions. 9 There examples in that report of the catalysts
is not designed for use in water and is not tested for biomimetic acylation of diols. We note that
there exist several methods for the monoacylation of diols utilizing approaches based on tinꢀ
10
containing reagents, borinic acids and organocatalysts. These methods are very effective but
none of these approaches utilizes a recyclable catalyst and are generally not useful in water. In
the present report we note the successful application of resinꢀbound lanthanides for the
biomimetic aminoacylation of nucleosides and nucleotides. The desired outcome in which the
materials promote the reaction and are easily recovered by filtration for use in further reaction
cycles has been successfully achieved.
Materials and Methods
Nuclear magnetic resonance analysis was performed at the CSICOMP Facility, University of
1
13
31
Toronto. NMR spectra were recorded at 300 MHz for H, 75 MHz for C, 121.5 MHz for P
and 42.5 MHz for 139La. Electrospray mass spectrometry analysis was performed at the AIMS
facility at the University of Toronto. Ion exchange resins (DOWEX™ HCRꢀW2, Amberlyst™
15, and Amberlite™ IRꢀ120) and lanthanide salts were purchased from commercial suppliers.
Metal ion resins were prepared by the literature procedure of Yu et. al. with the use of lanthanide
9
and magnesium triflates. Nꢀtꢀbocꢀphenylalanine ethyl phosphate (BocPheEP), benzoyl methyl
phosphate (BMP) and ethylene glycol monobenzoate were prepared by literature procedures. 1, 11
, 12
HPLC analysis for benzoylation reactions was performed on a C18 preparative column
(Waters ꢁBondapakTM 7.8 mm × 300 mm, 125 Å, 10 ꢁm) using an isocratic gradient with a
mobile phase of 10/90 (v/v) acetonitrile/water containing 0.1% trifluoroacetic acid (TFA) with a
flow rate of 1.5 mL/min and effluent detection at 230 nm. HPLC analysis for Nꢀtꢀboc
aminoacylation reactions was performed on C18 reversedꢀphase preparative column
(PhenomenexTM 10.00 mm x 250 mm, 90 Å, 10 ꢁm) with 40/60 (v/v) acetonitrile/water mobile
phase containing 0.1% TFA with a flow rate of 3.00 mL/min and effluent detection at 263 nm.
All acylation and aminoacylation reactions were conducted at room temperature in freshly
prepared HEPES buffer (0.1 M, pH = 8.0 at 23 °C).

Page 5 of 15
5
Monobenzoylation of ethylene glycol: Ethylene glycol (0.42 mmol) and LaIIIꢀDOWEX™ resin
(0.28 g resin, 0.42 mmol of LaIII) were added to a vial containing 3.0 mL of HEPES buffer. BMP
(0.42 mmol) was added to the vial and the mixture was stirred overnight. A 150 µL sample of the
reaction mixture was then subjected to HPLC analysis. The same reaction was performed with
LaIIIꢀ Amberlite™ IRꢀ120 resin in place of the LaIIIꢀDOWEX™ resin. Benzoic acid and glycol
monobenzoate were identified in the reaction by comparison to authentic materials. For catalyst
retrieval and recycling, the LaIIIꢀresin was isolated by filtration and washed with methanol and
water. The resin was then dried in an oven for two hours at 100 °C, cooled to room temperature
and weighed. The resin was added to another vial for another acylation reaction. The procedure
was repeated two more times in order to establish the recyclability of the catalyst.
Control experiments to establish reaction on the solid phase: 5 mL of freshly prepared
HEPES buffer was incubated with dried LaIIIꢀDOWEX™ resin and the mixture was stirred
overnight. The resin was filtered; 3.0 mL of the buffer was added to another vial containing
ethylene glycol and BMP (0.42 mmol each). The resulting mixture was stirred overnight
(reaction A). A parallel reaction was set up using 3.0 mL of fresh buffer that had not been
incubated with resin (reaction B). Each reaction was analyzed by HPLC. The filtered resin from
buffer incubation was washed with water and oven dried for two hours. One equivalent of the
resin was added to reaction A. The reaction was stirred overnight. As a control, LaIIIꢀDOWEX™
resin, which was not soaked in buffer, was added to reaction B. Each reaction was analyzed by
HPLC after stirring overnight.
Aminoacylation of nucleosides and nucleotides with solid phase lanthanides: Stock solutions
of nucleotides, nucleosides and BocPheEP, Mg(OTf)2 (80 mM each) and Ln(OTf)3 (160 mM)
were prepared in HEPES buffer. 100 mg of resin was used for reactions that were performed
with the solid phase catalyst. Reactions were conducted on a 1 mL scale. Substrate, lanthanide
salt (or resin), MgII (for nucleotide reactions) were added to a vial and made up to 800 µL by the
addition of buffer. The mixture was then stirred for five minutes. 200 µL of the BocPheEP stock
solution was added to the vial in order to initiate the reaction. The concentration of each
component in the reaction was 16 mmol except for the lanthanide salt, which was 32 mmol. The
reactions were stirred for 2 hours. Then, a 100 µL sample of the reaction mixture was subjected
to HPLC analysis.

Page 6 of 15
6
139La NMR Experiments: NMR experiments were performed on a Varian 400 MHz
spectrometer. A standard solution was prepared by dissolving LaCl3 in HEPES buffer. 350 µL of
this solution was diluted with an equal volume of D2O to give a final lanthanum concentration of
0.01 M. For analysis of the solid phase benzoylation reaction with LaIIIꢀDOWEX™ resin, 350
µL of the reaction solvent was transferred to an NMR tube and then diluted with an equal
volume of D2O.
Results and Discussion
benzoic
acid
BMP
glycol monoꢀ
benzoate
Figure 1. HPLC analysis of reactions that are based on lanthanideꢀimmobilized resin.
Both LaIIIꢀDOWEX™ and LaIIIꢀAmberlite™ IR120 resins promote monoacylation of
ethylene glycol (Figure 1). Based on the peak area of the glycol monobenzoate, we find that the
LaIIIꢀDOWEX is the more effective catalyst as it provided a slightly better yield based on HPLC
analysis. This finding is consistent with results reported by Yu et. al. where YbIIIꢀDOWEX™
9
was a more active catalyst compared to YbIIIꢀAmberlite™ IR120. The surface area for the
DOWEX™ resin is greater than that of the Amberlite™ IR120 resin and may account for the
13
observed difference in yield.
Control reactions with NaIꢀDOWEX™ in place of LaIIIꢀ
DOWEX™ provide no conversion of the acyl phosphate reagent into products (Figure 2). This
demonstrates that the lanthanide ion is essential and that neither the resin nor the sodium cation
alone promotes the reaction. The LaIIIꢀDOWEX™ resin from the acylation was isolated by
o
filtration, washed with water and dried at 100 C. The resin is then ready for further reactions.

Page 7 of 15
7
This was done for two further acylation reactions. In each case the HPLC product analysis was
identical to that with the initial catalyst.
BMP
Figure 2. HPLC analysis of acylation reaction with NaꢀDOWEX resin after 72 hours of reaction
time.
Reaction A
Reaction B
Figure 3. HPLC analysis for control experiments indicates no catalytic activity from leached
metal ions in solution. Reaction A condition: HEPES buffer (Laꢀresin incubation +), BMP,
ethylene glycol; Reaction B condition: HEPES buffer (Laꢀresin incubation –), BMP, and
ethylene glycol.

Page 8 of 15
8
Reaction B
+ La resin
Reaction A
+ La resin
Figure 4. HPLC analysis shows the addition of immobilized lanthanide resin promoted acylation
reaction in Reaction A and B. The chromatograms are comparable to those shown in Figure 1.
Figure 5. Left panel: 139La NMR spectra of 0.01 M LaCl3 standard in D2O. (Right) Aqueous
solvent from an acylation reaction after removal of the LaIIIꢀDOWEX™ resin.
Further control experiments were conducted in order to test whether the reactions occur with
14
the solid phase catalyst or with leached metal ions in solution.
We incubated the reaction
buffer with the catalytic resin overnight, removed the resin and tested the buffer for activity
(Reaction A). We find that the solution was inactive in promoting acylation, as the product
glycol benzoate peak was not observed in the chromatogram (Figure 3). Upon addition of the
immobilized catalyst, the reaction proceeded effectively (Figure 4). We then performed 139La
NMR experiments to see if any significant amounts of the metal ion were present in any of the
reaction mixtures after removal of the catalyst. There was no detectable signal in any of the

Page 9 of 15
9
mixtures we analyzed (Figure 5). This observation, in combination with the recyclability of the
catalyst, establishes that the reaction proceeds only in the presence of the solid phase catalyst.
Scheme 2. Aminoacylation of adenosine with BocPheEP using immobilized lanthanides.
In order to develop a solid phase method for tRNA acylation, it is necessary to demonstrate
that immobilized lanthanides are able to promote aminoacylation reactions. The reaction of
adenosine with BocPheEP was chosen as the starting point for solidꢀphase aminoacylation
reactions (Scheme 2). As the acylating agent is subject to competitive hydrolysis in the presence
of a lanthanide, an excess of the reagent can be used to ensure complete conversion of the
substrate. To evaluate the effectiveness of the solidꢀphase catalysts, reactions were carried out
with solidꢀphase resin (LaꢀDOWEX, LaꢀIR120, and NaꢀDOWEX as control). These reactions
were compared to those carried out with La(OTf)3. The reactions were then subject to HPLC
analysis and the resultant yield is reported as the total area of ester products. The HPLC product
peaks were isolated and their identity was confirmed as aminoacyl esters by mass analysis (ESIꢀ
MS(+) calculated m/z 514.2, found m/z 515.2 (M+H+)). Consistent with previous work from our
group, the lanthanide promoted reaction lead to formation of the 2’ꢀ and 3’ꢀaminoacyl esters of
adenosine with the 3’ꢀester being favored 15. Product analysis revealed that each lanthanide resin
was able to catalyze amino acylation of the nucleoside and the use of NaꢀDOWEX does to give
ester products (Figure 6). We find that the activity of the solidꢀphase catalyst is reduced in
comparison to the lanthanide salt La(OTf)3 as indicated by the reduction in the peak areas of the
ester products. This is common in the development of solidꢀphase methods and is attributed to
16
the diffusion of the reactants to the active metal ion within the resin matrix. However, the
differences are minimal and can be addressed by extension of the reaction time. Both gelꢀtype
resins (DOWEX™ and Amberlite™ IR120) are able to promote acylation with DOWEX™ again
being more effective. Amberlyst™ 15 is a macroporous resin that is specifically designed for

Page 10 of 15
10
17
catalysis due to its high surface area. We expected this resin to be the best choice for our
reaction. However, we observed that the macroporous resin was susceptible to mechanical
degradation when stirred. Therefore we chose DOWEX™ as the solid support for all subsequent
method development.
Figure 6. HPLC analysis for adenosine aminoacylation reactions. Adenosine is indicated in red
(peak 1) and the ester products are indicated in green (peak 2 & 3). Ester products are eluted in
the order of 2’ꢀ and 3’ꢀaminoacyl esters. (The ester products were identified by comparison of
retention times to previously reported results. 3)
We also assessed the effect of different immobilized lanthanide ions on the aminoacylation
reaction. We tested LaIII, EuIII YbIII, NdIII, DyIII, TbIII forms of the DOWEX™ catalyst. We tested
the ScIII form of the resin for comparison and the MgII form of the resin as a control. When
varying the lanthanide ion, highest yields were obtained with lanthanides with larger ionic radii

Page 11 of 15
11
(Table 1). This observation is consistent with the expected result based on lanthanide contraction.
18
Smaller lanthanides have a greater affinity for phosphates due to their charge density and are
not as efficient at coordination to adjacent hydroxyl groups in comparison to larger lanthanides.
Therefore, the optimal combination for biomimetic acylation reactions is provided by LaIIIꢀ
DOWEX™.
Table 1. Yields for the aminoacylation of adenosine with BocPheEP using differing DOWEXTM
catalysts.
Immobilized Metal Ion
Ionic Radius (pm) 18
Yield
28%
LaIII
NdIII
EuIII
TbIII
DyIII
YbIII
ScIII
MgII
103.2
98.3
94.7
92.3
91.2
86.8
74.5
72.0
26%
20%
20%
16%
8%
5%
No Conversion
With the optimized catalyst in hand, we then applied our optimized conditions to the
aminoacylation of cytidine and observed similar results. As an additional control, we performed
reactions with 2’ꢀdeoxy nucleosides as these substrates lack the required cisꢀdiol functionality to
coordinate effectively with the metal ion. HPLC analysis gave no indication of ester formation
(Figure 7).
The aminoacyl esters of adenosine are formed in a 2:1 ratio with the 3’ꢀisomer being
favoured. This is consistent with equilibrium ratios of oꢀacyl adenosine derivatives reported by
19
Griffin et. al.
We note that the selectivity of the aminoacylation reaction is not of great
concern as we intend to apply this method to the direct aminoacylation of tRNA. Chemically
modified tRNA can be used to incorporate unꢀnatural amino acids into proteins through in-vitro
translation methods. Peptide bond formation does not depend on the site of acylation on the
terminal ribonucleotide of tRNA as the resultant ester is subject to rapid equilibration between

Page 12 of 15
12
20
21
the 2’ and 3’ positions. We have observed similar behaviour in previously reported work.
Selectivity is of greater importance in applications where the site of acylation needs to more
precise e.g. regioselective protection of carbohydrates. 22
2.2
2
1.8
1.6
2.2
2
1.8
1.6
1.4
1.4
1.2
1.2
1
0.8
0.6
1
0.8
0.6
0.4
0.2
0
0.4
0.2
0
-0.2
-0.2
0
2
4
6
8
10
12
14
0
2
4
6
8
10 12 14
Time (Min)
Time (min)
Figure 7. (Left) aminoacylation of adenosine with LaIIIꢀDOWEX™ resin. (Right)
aminoacylation of 2’ꢀdeoxyadenosine under the same conditions, no esters formed.
2.2
2
1.8
1.6
1.4
1.2
1
2.2
2
1.8
1.6
1.4
1.2
1
AMP
AMP
ester
0.8
0.6
0.4
0.2
0
0.8
0.6
0.4
0.2
0
0
20
40
0
20
40
Time (Min)
Time (Min)
Figure 8. HPLC output for aminoacylation of 5’ꢀAMP with use of LaIIIꢀDOWEX with (left) and
without added MgII (right).
In contrast to reactions of nucleosides, additional complexity is provided by nucleotide
substrates whose phosphate groups may bind and remove lanthanum from the solid support. It
has been shown that MgII can be introduced into the reaction mixture in order to bind to
phosphate groups, allowing the lanthanide to be available for catalysis. Therefore, the addition of
MgII into the solidꢀphase reaction would be ideal for the reaction of nucleotide substrates. We

Page 13 of 15
13
performed reactions with adenosineꢀ5’ꢀmonophosphate (5’ꢀAMP) and the immobilized catalyst.
We noted ester formation takes place in the presence of added MgII (Figure 8).
The newly formed peak was isolated and analyzed by ESIꢀMS and we confirmed that the
peak is an aminoacyl ester of the nucleotide (ESIꢀMS (ꢀ) calculated m/z 594.2, found m/z 593.2
(MꢀH+)). Similar results were obtained in the acylation of cytidine 5’ꢀmonophosphate and
uridine 5’ꢀmonophosphate.
Conclusions
Lanthanides immobilized onto ionꢀexchange resins effectively catalyze monoacylation of
ethylene glycol in addition to the aminoacylation of nucleosides and nucleotides. The
development of the catalysts greatly improves the efficiency of the reaction as the catalysts are
easily recovered by filtration and reused for further reactions. With this development, we are able
to reuse the lanthanide which was originally discarded when used as a homogenous catalyst.
Finally, the addition of MgII to the reaction mixture improves the efficiency of acylation when a
ribonucleotide serves as a substrate. Based on these results, we will be expanding the scope of
this method to include carbohydrate acylation and the direct aminoacylation of tRNA.
Acknowledgements
We thank NSERC Canada for support through a Discovery Grant. We also thank Dr. Darcy
Burns for valuable assistance and advice with 139La NMR experiments.
References:
1.
Kluger, R.; Cameron, L. L., J Am Chem Soc 2002, 124 (13), 3303ꢀ3308; Cameron, L. L.;
Wang, S. C.; Kluger, R., J Am Chem Soc 2004, 126 (34), 10721ꢀ10726.
2.
Gray, I.; Ren, R.; Westermann, B.; Kluger, R., Can. J. Chem. 2006, 84 (4), 620ꢀ624;
Gray, I. J.; Kluger, R., Carbohydr Res 2007, 342 (14), 1998ꢀ2002.
3.
4.
Tzvetkova, S.; Kluger, R., J Am Chem Soc 2007, 129 (51), 15848ꢀ15854.
Duffy, N. H.; Dougherty, D. A., Org Lett 2010, 12 (17), 3776ꢀ3779.

Page 14 of 15
14
5.
Kobayashi, S.; Sugiura, M.; Kitagawa, H.; Lam, W. W., Chem Rev 2002, 102 (6), 2227ꢀ
2302; Barrett, A.; Braddock, D., Chem Commun 1997, (4), 351ꢀ352; Kobayashi, S.; Hashiya, I.;
Takahori, T.; Araki, M.; Ishitani, H., Tetr Lett 1992, 33 (45), 6815ꢀ6818; Kobayashi, S.;
Nagayama, S.; Busujima, T., Journal of the American Chemical Society 1998, 120 (32), 8287ꢀ
8288.
6.
Angyal, S., Chem Soc Rev 1980, 9 (4), 415ꢀ428; Angyal, S.; Mills, J., Aust J Chem 1985,
38 (8), 1279ꢀ1285; Angyal, S.; Craig, D., Carbohydr Res 1993, 241, 1ꢀ8; Angyal, S., Aust J
Chem 2002, 55 (1ꢀ2), 79ꢀ81; Angyal, S., Aust J Chem 2005, 58 (1), 58ꢀ59.
7.
Chesney, A., Green Chem 1999, 1 (5), 209ꢀ219; Anastas, P.; Eghbali, N., Chem. Soc.
Rev. 2010, 39 (1), 301ꢀ312.
8.
Kobayashi, S.; Nagayama, S., J Am Chem Soc 1998, 120 (12), 2985ꢀ2986; Nagayama, S.;
Kobayashi, S., Angew Chem Int Ed 2000, 39 (3), 567ꢀ569.
9.
Yu, L.; Chen, D.; Li, J.; Wang, P., J Org Chem 1997, 62 (11), 3575ꢀ3581.
10.
Lee, D.; Taylor, M., J Am Chem Soc 2011, 133 (11), 3724ꢀ3727; Kawabata, T.;
Muramatsu, W.; Nishio, T.; Shibata, T.; Schedel, H., J Am Chem Soc 2007, 129 (42), 12890ꢀ
12895; Muramatsu, W.; William, J.; Onomura, O., J Org Chem 2012, 77 (1), 754ꢀ759; Demizu,
Y.; Kubo, Y.; Miyoshi, H.; Maki, T.; Matsumura, Y.; Moriyama, N.; Onomura, O., Org Lett
2008, 10 (21), 5075ꢀ5077; Clarke, P.; Kayaleh, N.; Smith, M.; Baker, J.; Bird, S.; Chan, C., J
Org Chem 2002, 67 (15), 5226ꢀ5231.
11.
Kluger, R.; Li, X.; Loo, R., Can J Chem 1996, 74 (12), 2395ꢀ2400.
12.
Iwasaki, F.; Maki, T.; Onomura, O.; Nakashima, W.; Matsumura, Y., J Org Chem 2000,
65 (4), 996ꢀ1002.
13.
Sarma, D.; Kumar, A., Appl Catal A 2008, 335 (1), 1ꢀ6; Izci, A.; Bodur, F., React Funct
Polym 2007, 67 (12), 1458ꢀ1464.
14.
Davies, I.; Matty, L.; Hughes, D.; Reider, P., J Am Chem Soc 2001, 123 (41), 10139ꢀ
10140; Hudson, R.; Li, C.; Moores, A., Green Chem 2012, 14 (3), 622ꢀ624.
15.
16.
17.
18.
Tzvetkova, S.; Kluger, R., J Am Chem Soc 2007, 129 (51), 15848ꢀ15854.
Barbaro, P.; Liguori, F., Chem Rev 2009, 109 (2), 515ꢀ529.
Harmer, M.; Sun, Q., Appl Catal A 2001, 221 (1ꢀ2), 45ꢀ62.
Greenwood, N. N.; Earnshaw, A., Chemistry of the Elements. 2nd Edition ed.;
Butterworth–Heinemann: Oxford, 1998.

Page 15 of 15
15
19.
Griffin, B.; Jarman, M.; Reese, C.; Sulston, J.; Trentham, D., Biochemistry 1966, 5 (11),
3638ꢀ3649.
20.
21.
Rangelov, M.; Vayssilov, G.; Petkov, D., Int J Quantum Chem 2006, 106 (6), 1346ꢀ1356.
Her, S.; Kluger, R., Org Biomol Chem 2011, 9 (3), 676ꢀ678.
22.
Codee, J.; Ali, A.; Overkleeft, H.; van der Marel, G., Compt Rend Chim 2011, 14 (2ꢀ3),
178ꢀ193.
Graphical Abstract