10
L. Li et al. / Catalysis Communications 64 (2015) 6–11
templates, so carbon content was halved, while oxygen and silicon
increased sharply (L vs. Z, Table 2). Binding energy and atomic percent-
age from L to CL then GH-CL demonstrated that silica reagent and gua-
nosine were attached to the surface of L, a similar tendency was found
on L to EL then SG-EL (Table 2). Based on nitrogen content, loading of
guanosine had an order: GH-CL N GH-CD N GH-C (Table 2), reflecting in-
fluence of sodium lactate too.
whose role was similar to the secondary amine-catalyzed hetero-Diels–
Alder [7]. After immobilization, acidified guanosine promoted
e.e. (entries 7, 8, 12 vs. 5), probably because guanosine was confined
in a more rigid environment. Recycling demonstrated that GH-CL
degrades slowly, but major configuration of product was maintained
(entry 8).
L and D were both internally right-handed while Z left-handed, but a
+
−
Sample Z was composed of twisted rods having a length of 0.2–1 μm
combination of [GH] Cl with CL was more enantioselective than that
with CD or Z (entries 8 vs. 12 and 7). In addition, the same level of e.e.
and lower conversion were obtained in heptaldehyde and ethyl
glyoxylate (entries 10 and 11), and much reduced results were obtained
in 4-chlorobenzaldehyde (entry 9), but acetophenone was completely
inert, illustrating that aldehyde was more active than ketone [1,7].
Condensation of guanosine with salicylaldehyde gave 65% conver-
(
Z, Fig. 4). Addition of chiral sodium lactate produced longer helical
rods, where L showed length of 0.2–3 μm, pitch of 0.5–1.5 μm, as
well as offset (distance from spiral axis to center of rod) of 50–250 nm
(
L vs. Z). L and D had both right- and left-handed morphology, demon-
strating that sodium lactate may not work with morphology (Fig. 4).
GH-CL, SG-EL and SG-DL degraded by different degrees perhaps due to
hydrolysis in synthesis (Fig. 4, Section 2.2.2) [20]. SG-EL looked more
helical than GH-CL and SG-DL, but became bolder and longer than L
+
−
sion as well as 39% e.e., higher than guanosine and [GH] Cl (entries
13 vs. 4 and 5). In view of TADDOL [24], salicylaldehyde might provide
an intramolecular hydrogen bond between phenolic oxygen with
hydroxyl of guanosine, then intermolecular hydrogen bond between
benzaldehyde and guanosine looked more exclusive leading to im-
proved stereoselectivity (Scheme 1).
(Fig. 4). TEM confirmed that Z, L, GH-CL and SG-EL were solid rods in-
stead of hollow tubes (Fig. S4). Based on enantioselective adsorption
of valine (Fig. S5), Z preferred L-valine more than D-valine, indicating
excessive left-handed channels [16]. But both L- and D-sodium lactate
changed adsorption priority of Z, which meant that L and D were inter-
nally righted-handed [16].
Immobilization of SG improved enantioselectivity, because configu-
ration of SG was confined in channels of E, EL or ED (entries 15, 16, 20
vs. 13). SG-E showed a conversion less than half of SG-EL and SG-ED
(entries 15 vs. 16 and 20), due to its poor surface area (Table 1).
SG-EL provided better e.e. than SG-ED (entries 16 vs. 20), probably
because L showed a much larger adsorption difference between L-
and D-sodium lactates than D (Fig. S5). Recycling of SG-EL was satis-
factory during six cycles that witnessed that degradation was not
very sharp in dichloromethane (entry 16). Heptaldehyde and ethyl
glyoxylate were still moderate substrates (entries 18 and 19).
4-Chlorobenzaldehyde was not as good as benzaldehyde either
(entries 17 vs. 16), perhaps because electron-withdrawing of chlorine
3
.2. Catalysis
Catalyst blank showed 3% conversion, indicating that molecular
sieve (4 Å) was catalytically active to some extent (entry 1, Table 3),
which also proved that sulfonated silica (CL) was inactive (entry 2). EL
improved conversion perhaps owing to ammoniums, while e.e. of 9%
meant that channels of EL were enantioselective (entry 3).
Acidified guanosine showed improved conversion and enantio-
selectivity than guanosine (entries 5 vs. 4), perhaps owing to ammonium,
Table 3
Asymmetric hDA reaction of Danishefsky's diene with aldehydes catalyzed by guanosine derivatives.
Entrya
Catalyst
Aldehyde (R)b
Conversionc (%)
E.e.d (%)
TOFe (h−1
)
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
None
CL
EL
Ph
Ph
Ph
Ph
Ph
p-ClPh
Ph
Ph
p-ClPh
3
2
6
10
39
6
37
0
0
–
−
−
−
2
1.6 × 10
5.0 × 10
8.3 × 10
3.2 × 10
5.0 × 10
3.0 × 10
4.5 × 10
8.3 × 10
2.5 × 10
2.5 × 10
4.7 × 10
5.4 × 10
2.2 × 10
1.6 × 10
5.9 × 10
2.5 × 10
3.7 × 10
2.1 × 10
4.2 × 10
2
9 (R)
7 (R)
11 (R)
12
19 (R)
33 (39,35,20,22,27) (R)
15
45
32
16 (R)
39 (R)
30
59 (R)
61 (60,50,57,63,47) (R)
40
29
2
Guanosine
+
−
−
−1
−2
[GH] Cl
+
[GH] Cl
−
−
−
−
−
−
−
−
−
−
−
−
−
−
1
1
2
1
1
1
1
1
1
1
1
1
1
1
GH-C
GH-CL
GH-CL
GH-CL
GH-CL
GH-CD
SG
55 (51,55,49,45,30)
10
30
31
57
65
27
20
0
1
2
3
4
5
6
7
8
9
0
n-C
6
H
CH
13
CH
Ph
Ph
p-ClPh
Ph
Ph
p-ClPh
n-C
CH
Ph
3
2
OC_O
SG
SG-E
SG-EL
SG-EL
SG-EL
SG-EL
SG-ED
71 (70,73,66,50,40)
31
45
26
51
H
6 13
3
CH
2
OC_O
37
55 (R)
a
2 2
Conditions: Danishefsky's diene (1.0 mmol), aldehyde (1.0 mmol), catalyst (20 mol% of N based on diene), MS 4 Å (50 mg), and CH Cl (3 mL). For entry 2, catalyst (20 mol% of Na
based on diene).
b
Benzaldehyde, 4-chlorobenzaldehyde, heptaldehyde, and ethyl glyoxylate as four substrates.
Molar ratio of product to original diene, values in parenthesis represented recycling data.
Enantiomeric excess. Retention time (min): product of benzaldehyde, 3.570 (R) and 3.880 (S), configuration determined by comparison of retention time from literature [21]; product
c
d
of 4-chlorobenzaldehyde, 3.598 (major) and 3.918 (minor) [21]; product of heptaldehyde, 3.942 (major) and 7.289 (minor) [21,22]; product of ethyl glyoxylate, 3.944 (major) and 7.399
minor) [23], no absolute configuration determined for three substrates.
(
e
−1
−1
Turnover frequency of cycle fresh, molproductmol(N or Na) (6 h)
.