Reactivity of halo sugars on silver cathodes

Article abstract of DOI:10.1135/cccc20000881

Our extensive research on the reductive electrochemistry of glycosyl derivatives has revealed a remarkable electrocatalytic activity of silver cathodes in the reduction of glycosyl halides, which is strictly related to both the electrode active surface and the reaction medium. The peculiarities exhibited by silver cathodes have been extensively characterised (i) by reactivity studies performed on a wide range of halo sugars by cyclic voltammetry and (ii) by preparative electrolyses leading to the formation of simple and mixed glycosyl dimers and, in the presence of suitable acceptors, of glycosides and glycosyl compounds.

Full text of DOI:10.1135/cccc20000881

Reactivity of Halo Sugars
881
REACTIVITY OF HALO SUGARS ON SILVER CATHODES
a3
Sandra B. RONDININIa1, Patrizia R. MUSSINIa2,*, Federico CRIPPA
Michela PETRONE and Guido SELLO
,
a
b
a Department of Physical Chemistry and Electrochemistry, University of Milano,
Via Golgi 19, 20133 Milano, Italy; e-mail: 1 sandra.rondinini@unimi.it,
2 pmussini@mailserver.unimi.it, 3 crippafed@tin.it
b Department of Organic and Industrial Chemistry, University of Milano, Via Venezian 21,
20133 Milano, Italy; e-mail: sellogui@icil64.cilea.it
Received October 13, 1999
Accepted Jan uary 9, 2000
Presented at the 32nd Heyrovský Discussion on Organic Electrochemistry, Třeš3, June 6–10, 1999.
Our exten sive research on th e reductive electroch em istry of glycosyl derivatives h as revealed
a rem arkable electrocatalytic activity of silver cath odes in th e reduction of glycosyl h alides,
wh ich is strictly related to both th e electrode active surface an d th e reaction m edium . Th e
peculiarities exh ibited by silver cath odes h ave been exten sively ch aracterised (i) by reactivity
studies perform ed on a wide ran ge of h alo sugars by cyclic voltam m etry an d (ii) by prepara-
tive electrolyses leadin g to th e form ation of sim ple an d m ixed glycosyl dim ers an d, in th e
presen ce of suitable acceptors, of glycosides an d glycosyl com poun ds.
Key w ord s: Silver cath ode; Halo sugars; Glycosyl dim ers; Glycosides; Electrocatalytic activ-
ity; Electrosyn th esis; Electroreduction s; Preparative electroch em istry.
Quite recen tly, our exten sive research work in th e field of th e reductive
electroch em istry of glycosyl derivatives1–5 h as revealed a rem arkable
electrocatalytic activity of silver cath odes in th e reduction of glycosyl
h alides. Actually silver turn ed out5 to be th e cath ode m aterial on wh ich th e
above reaction takes place at th e lowest cath odic poten tial, very probably
on accoun t of th e powerful silver–h alide in teraction actin g as a drivin g
force for leavin g group extraction . Moreover, silver turn ed out to be th e
best cath ode for preparative electrosyn th eses of in terestin g products, sev-
eral of th em bein g quite un expected con siderin g usual reaction path ways in
organ ic ch em istry, such as glycosyl dim ers in h igh , n early statistical
yields2–5, an d aryl glycosides wh en perform ed in th e presen ce of ph e-
n ols2,3,5. Th ese results were in terpreted in term s of radical couplin g4,5, th e
silver cath ode providin g a sort of “cage effect”.
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

882
Rondinini et al.:
Th ese in terestin g features prom pted a deeper, exten sive in vestigation of
electrocatalytic properties of silver as a fun ction of two m odulatin g factors,
n otably th e electrode active surface an d th e reaction m edium . Th ese studies
provide a powerful basis for correlation an d ration alisation of previous re-
sults con cern in g glycosyl h alide/silver system s. Th e presen t work th erefore
m erges our latest studies on electrocatalytic properties of silver with a re-
view/system atisation of th e m ost relevan t previous results.
EXPERIMENTAL
Cyclic voltammetry experiments. Voltam m etric in vestigation s of a variety of h alo sugars
were carried out eith er with an AMEL 466 polarograph ic an alyser, or with a station in clud-
in g an AMEL 553 poten tiostat/galvan ostat, an AMEL 566 fun ction gen erator an d an AMEL
862a xy recorder, at 200 or 100 m V s–1 sweep rate, in HPLC-grade aceton itrile (ACN), with a
wide ran ge of supportin g electrolytes, an d with an aqueous saturated calom el electrode
(SCE) as a referen ce electrode.
Ag workin g electrodes con sisted of differen tly sh aped platin um supports (wires 0.5 m m in
diam eter an d of 2–6 m m len gth ; a disk 0.5 m m in diam eter; a h em isph ere 0.6 m m in diam e-
ter) covered with silver by cath odic deposition again st a silver an ode from a 10 g dm –3
KAg(CN)2 bath , th e deposition curren ts ran gin g from 0.5 to 5 m A cm –2 (th e lower value be-
in g th e stan dard curren t den sity used in th e preparation of th e silver layer of silver ch loride
electrodes for th erm odyn am ic experim en ts7). Th e rough n ess factor8 was determ in ed (i) by
un derpoten tial deposition 9 of lead in th e poten tial ran ge from –0.600 to –0.750 V vs SCE at
5 m V s–1 scan rate, with 280 µC cm –2 specific ch arge factor10, an d (ii) by capacitan ce deter-
m in ation in th e poten tial ran ge from –0.200 to –0.400 V vs SCE, with 50 µF cm –2 as specific
capacitan ce factor10
.
Preparative electrolysis experiments. Preparative electrolyses were carried out in a two-
com partm en t cell. Th e cath ode con sisted of a 1 × 10 × 0.1 cm rectan gular Pt plate, electro-
lytically covered with silver as described above. A silver plate in ACN saturated with
tetraeth ylam m on ium brom ide (TEABr) was used as an ode, to provide a h arm less coun ter-
reaction (n am ely, Ag + Br– → AgBr + e). An an ion -exch an ge m em bran e (Sybron MA3475)
or a glass frit were used to separate th e cath ode an d an ode com partm en ts. An AMEL 551
poten tiostat/galvan ostat was used to run electrolyses un der both poten tiostatic or
galvan ostatic con trol.
Com m ercial acetobrom oglucose (Fluka, purum 95%) stabilised by calcium carbon ate, was
purified by filtration an d recrystallisation from dieth yl eth er.
Direct-ph ase TLC (h exan e/eth yl acetate or propan -2-ol/CH2Cl2) were used to m on itor th e
progress of th e electrolytic processes. Reaction products were isolated by direct-ph ase flash
ch rom atograph y (h exan e/eth yl acetate or propan -2-ol/CH2Cl2) an d iden tified by m ean s of
NMR an d MS as described in previous papers1–5
.
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

Reactivity of Halo Sugars
883
RESULTS AND DISCUSSION
Voltammetric Experiments
Effect of Cath ode Nature
Tetra-O-acetyl-α-D-glucopyran osyl brom ide (acetobrom oglucose (1)) was
ch osen as a m odel for testin g an om eric glycosyl h alide reactivity on a n um -
ber of m etallic cath ode m aterials, in cludin g silver, palladium , platin um ,
m ercury an d zin c5. Silver was foun d to exh ibit th e best electrocatalytic ac-
tivity for acetobrom oglucose reduction ; acetobrom oglucose reduction peak
poten tials on silver are less n egative by 600–700 m V th an on m ercury,
wh ich in turn is m uch m ore favourable to acetobrom oglucose reduction
th an a very popular cath ode m aterial as glassy carbon 11, h avin g n o specific
in teraction s at all with h alide ion s.
Effect of Rin g Con figuration
Table I provides a system atised syn opsis of th e reduction peak poten tials
on silver an d/or m ercury for all glycosyl h alides h ith erto in vestigated1–5
(Fig. 1).
AcO
AcO
CH2OAc
O
AcOH2C
AcO
AcO
CH2OAc
O
OAc
O
AcOH2C
AcO
AcO
O
AcO
AcO
CH2OAc
O
OAc
OAc
OAc
O
AcO
1
Br
2
Br
3
Br
4
OAc
OAc
I
Br
AcO
O
BzOH2C
BzO
BzO
CH3
CH3
O
O
O
O
AcO
OAc
O
OBz
OAc
Br
AcO
Br
O
5
Br
6a
6b
O
7
BenzOH2C
BenzO
BenzO
AcOH2C
AcO
AcO
O
AcOH2C OAc
O
O
O
O
AcO
O
AcO
O
O
HO
I
CH2I
Cl
I
O
10
11
8
9
FIG. 1
Structures of com poun ds ch aracterised by cyclic voltam m etry (Table I)
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

884
Com parin g an om eric brom ides it is eviden t th at:
Rondinini et al.:
a) th eir reactivity is in fluen ced by configuration of th e carbon atom adja-
cen t to th e leavin g group (m an n ose, h avin g th e large acetate protective
group trans in stead of cis, an ticipates glucose by 0.3 V on m ercury an d by
0.1 V on silver). On th e con trary, in version of con figuration (galactose)
as well as a bulky substituen t (m altose) at th e C4 carbon are n early with out
effect;
b) enantiomers h ave th e sam e reactivity on each cath ode m aterial, as ex-
pected;
TABLE I
Glycosyl h alide reduction peak poten tials, (Ep, V vs SCE) obtain ed in 0.1 M aceton itrile/TEAP,
at 200 m V s–1 scan rate
Brom ides
Hg
Ag
Tetra-O-acetyl-α-D-glucopyran osyl brom ide 1
Tetra-O-acetyl-α-D-m an n opyran osyl brom ide 2
Tetra-O-acetyl-α-D-galactopyran osyl brom ide 3
Tetra-O-acetyl-α-D-m altopyran osyl brom ide 4
Tetra-O-ben zoyl-α-D-glucopyran osyl brom ide 5
–1.90
–1.60
–1.90
–1.95
–1.25 –1.65a
–1.17 –1.55
–1.25
–1.35 –1.7a
–1.72
–1.25a –1.55
–2.13b –2.28b
–2.17b –2.27b
Tri-O-acetyl-α-D-fucopyran osyl brom ide 6a
Tri-O-acetyl-α-L-fucopyran osyl brom ide 6b
–1.95
–1.95
–1.20
–1.20
Iodides
Hg
Ag
6-Deoxy-6-iodo-1,2:3,4-di-O-isopropyliden e-
α-D-galactopyran ose 7
–1.81
–1.23
(3,4,6-Tri-O-ben zyl-α-D-glucosyl)iodom eth an e 8
–1.435
–1.20
–1.23
–1.05
Isopropyl 2-deoxy-2-iodo-3,4,6-tri-O-acetyl-
D-glucopyran oside 9
3-Deoxy-1,2:5,6-di-O-isopropyliden e-3-iodo-
α-D-glucofuran oside 10
–1.45
Hg
–1.23
Ch loride
Ag
Tetra-O-acetyl-α-D-m an n opyran osyl ch loride 11
–2.65a
a
b
Sh oulder; ben zoate peak.
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

Reactivity of Halo Sugars
885
c) ben zoate protecting groups, yet m uch larger th an acetates, im prove th e
glycosyl h alide reactivity on m ercury probably on accoun t of th eir greater
ability to stabilise th e in term ediate. Clearly, th ey are m ore reactive th an ac-
etates, givin g reduction peaks of th eir own , wh ich suggests a possibility of
in -site sugar deprotection .
Com parin g brom ides with oth er h alides, n otwith stan din g th e differen ces
arisin g from eith er th e sugar con figuration or n ature of protectin g groups
(steric h in dran ce, in term ediate stabilisation effects), silver sh ows a rem ark-
able levellin g effect sin ce both glycosyl brom ides an d iodides (with on ly
on e exception ) reduce in th e sam e n arrow poten tial ran ge (ca from –1.2 to
–1.25 V vs SCE), wh ile th e reactivity sequen ce I > Br is observed on m er-
cury.
Effect of Active Surface
Figure 2 sh ows th e relation sh ip between th e curren t den sity for silver depo-
sition an d th e rough n ess factor, determ in ed by both un derpoten tial deposi-
tion an d capacitan ce for several sh ort-wire electrodes. Th e results are quite
reproducible, th ough capacity-active surfaces are con stan tly larger th an
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
0.00
0.50
1.00
1.50
2.00
2.50
3.00
–2
j, mA cm
FIG. 2
Rough n ess factors determ in ed by un derpoten tial deposition (full sym bols) an d capacity
(em pty sym bols) as a fun ction of th e silver deposition curren t, for sh ort-wire electrodes
(geom etric surface in cm 2: 0.037 (trian gles), 0.094 (squares), 0.110 (circles))
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

886
Rondinini et al.:
un derpoten tial deposition on es (wh ich is n ot un expected, as th e two crite-
ria rely on differen t “probes”).
In creasin g th e silver electrode active surface results in a rem arkable posi-
tive sh ift of first-cycle reduction peak poten tials, as it is eviden t from Fig. 3
(n early +200 m V sh ift in creasin g rough n ess by a factor of ten ).
It is rem arkable th at th e first-cycle peak at –1.25 V vs SCE was observed to
abruptly decrease with in a few subsequen t cycles, wh ile th e sign al at
–1.75/–1.8 V in creased, becom in g a well defin ed peak of h eigh t com parable
with th e in itial peak at –1.25 V. Neverth eless, a sh ort an odic pulse or,
better, a redox activation cycle in cludin g th e poten tial of th e reversible
couple Ag/AgBr allowed easy restoration of th e first-cycle situation . Figure 4
sh ows a typical pattern obtain ed with tetraeth ylam m on ium perch lorate
(TEAP) as supportin g electrolyte, a con dition im plyin g th at th e brom ide in -
volved in th e AgBr reversible wave results from acetobrom oglucose reduc-
tion . Th is suggests th at som e of th e reduction products are able to film over
th e electrode, an d th at th e film ed-over surface loses m uch of its catalytic
activity. Th us th e two reduction peaks could be ascribed to “active” an d
“film ed” silver surface.
Note th at th e film -over action can n ot be ascribed to th e brom ide ion it-
self, th ough eviden tly rem ain in g on th e electrode surface, as workin g with
–1.35
–1.30
–1.25
–1.20
–1.15
0.00
0.50
1.00
1.50
2.00
2.50
–2
j, mA cm
FIG. 3
Acetobrom oglucose reduction peak poten tials on electrolytically prepared silver cath odes of
geom etric surface 0.0377 cm 2 (ꢀ) an d 0.0659 cm 2 (ꢀ), as a fun ction of th e silver deposition
curren t j
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

Reactivity of Halo Sugars
887
con cen trated brom ide supportin g electrolytes results in ca –200 m V sh ift
on ly (see below).
Effect of Supportin g Electrolyte
Th e effect of supportin g electrolyte h as been tested by an exten sive sch em e
of cyclic voltam m etry experim en ts, workin g with tetram eth ylam m on ium
perch lorate (TMAP), tetrafluoroborate (TMATFB), ch loride (TMACl), an d io-
dide (TMAI); tetraeth ylam m on ium perch lorate (TEAP), tetrafluoroborate
(TEATFB), brom ide (TEABr), an d iodide (TEAI); tetrabutylam m on ium per-
ch lorate (TBAP), tetrafluoroborate (TBATFB), an d iodide (TBAI);
tetrah eptylam m on ium brom ide (THepABr) an d iodide (THepAI).
Th e results are sum m arised in Table II an d visualised in Figs 5a an d 5b
(wh ere, for com parison , all curren ts h ave been n orm alised at 100 m V s–1 on
th e basis of th eir depen den ce on th e square root of poten tial sweep rate).
Peak potentials: Acetobrom oglucose reduction poten tials are lowest in per-
ch lorate, ch loride, an d tetrafluoroborate solution s, m ore n egative in bro-
m ide solution s, an d even m ore n egative in iodide solution s.
Peak currents (see Figs 5a an d 5b):
A) With in each an ion series, acetobrom oglucose reduction curren ts are
regularly m odulated by th e alkyl ch ain len gth of th e quatern ary am m o-
a
c
d
0.2
I, mA
b
0.1
d
0.0
–0.5
–1.0
–1.5
–2.0
E, V vs SCE
FIG. 4
Typical cyclovoltam m etric pattern for 0.003 M acetobrom oglucose in aceton itrile with 0.1
M
TEAP (or TEATFB) as a supportin g electrolyte on a silver-wire cath ode (geom etric surface
0.16 cm 2, electrodeposition curren t den sity 0.5 m A cm –2); a first cycle, b in term ediate cycle,
c station ary cycle, d after a few reactivation cycles in cludin g th e Ag/AgBr reversible wave
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

888
Rondinini et al.:
TABLE II
Cyclic voltam m etry (poten tial ran ge –0.6 to –2.2 V vs SCE, scan rate 200 m V s–1) for
acetobrom oglucose (1) reduction on fresh silver-wire electrodes prepared at j = 0.5 m A cm –2
Ip, m A
1st cycle
Ip, m A
2n d cycle
1, m ol dm –3
Ep1, V vs SCE
Medium
Tetram eth ylam m on ium salts
Aceton itrile/0.1M TMAP
0.003
0.006
0.009
0.012
0.015
0.015
0.015
–1.275
–1.300
–1.350
–1.375
–1.425
–1.425
–1.500
0.112
0.246
0.328
0.400
0.496
0.404
0.364
0.090
0.178
0.220
0.264
0.372
0.304
0.240
+ 0.01 m ol dm –3 TBATFB
+ 0.1 m ol dm –3 TBATFB
Aceton itrile/0.1 M TMATFB
0.003
0.006
0.009
0.015
0.015
0.015
–1.275
–1.300
–1.350
–1.400
–1.400
–1.350
0.226
0.382
0.532
0.84
0.146
0.228
0.312
0.532
0.600
0.484
+ 0.01 m ol dm –3 TBATFB
+ 0.1 m ol dm –3 TBATFB
0.812
0.748
Aceton itrile/TMACl sat.a
0.01
0.02
0.03
–1.275
–1.325
–1.350
0.078
0.134
0.190
0.050
0.090
0.120
Aceton itrile/TMAI sat.a
0.01
0.02
0.03
–1.450
–1.475
–1.525
0.096
0.178
0.256
0.070
0.128
0.186
Tetraeth ylam m on ium salts
Aceton itrile/0.1 M TEAP
0.001
0.002
0.003
0.003
0.003
0.003
–1.200
–1.225
–1.225
–1.425
–1.400
–1.400
0.112
0.218
0.318
0.312
0.306
0.268
0.070
0.156
0.222
0.226
0.236
0.190
+ 0.1 m ol dm –3 TEABr
+ 0.01 m ol dm –3 TBAP
+ 0.1 m ol dm –3 TBAP
Aceton itrile/0.1 M TEAPa
0.001
0.002
–1.275
–1.275
0.086
0.174
0.074
0.112
Aceton itrile/0.1 M TEATFBa
0.001
0.002
0.003
–1.250
–1.275
–1.300
0.106
0.190
0.216
0.054
0.120
0.190
Aceton itrile/0.1 M TEABra
0.001
0.002
0.003
–1.350
–1.65
0.096
0.184
0.264
0.062
0.116
0.164
–1.380
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

Reactivity of Halo Sugars
889
TABLE II
(Continued)
Ip, m A
1st cycle
Ip, m A
2n d cycle
1, m ol dm –3
Ep1, V vs SCE
Medium
Aceton itrile/0.1 M TEAIa
0.001
0.002
0.003
–1.565
–1.550
–1.525
0.098
0.202
0.298
Tetrabutylam m on ium salts
Aceton itrile/0.1 M TBAP
0.001
0.002
0.003
0.004
0.005
n o peak
n o peak
n o peak
n o peak
n o peak
Aceton itrile/0.1 M TBATFB
0.003
0.006
0.009
0.012
0.015
0.024
0.024
–1.300
–1.350
–1.400
–1.450
–1.450
–1.450
–1.450
0.038
0.080
0.170
0.142
0.180
1.360
0.036
0.056
0.112
0.110
0.148
0.920
+ 0.1 m ol dm –3 TEABr
Aceton itrile/0.1 M TBAIa
0.001
0.002
0.003
–1.525
–1.525
–1.525
0.084
0.168
0.236
0.062
0.130
0.172
Tetrah eptylam m on ium salts
Aceton itrile/0.1 M THepABr
Aceton itrile/0.1 M THepAI
0.003
0.006
0.009
0.015
–1.400
–1.425
–1.425
–1.425
0.288
0.444
0.712
0.96
0.16
0.29
0.436
0.72
0.001
–1.400
0.076
0.046
0.002
0.003
–1.425
–1.430
0.128
0.184
0.76
0.112
a
Scan rate 100 m V/s.
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

890
Rondinini et al.:
n ium ion , with a m axim um con stan tly correspon din g to th e tetraeth yl-
am m on ium cation . An yway, th e cation m odulation is m uch m ore eviden t
with an ion s h avin g little affin ity for th e silver cath ode, such as tetrafluoro-
borates an d perch lorates.
B) In th e tetrafluoroborate an d perch lorate series, givin g th e m axim um
curren t decrease with in creasin g len gth of alkyl ch ain , th e tetrabutyl-
am m on ium cation is sufficien t to n early (TBATFB) or totally (TBAP) preven t
acetobrom oglucose reduction , wh ile even lon ger-ch ain h alide supportin g
electrolytes result in fairly h igh curren ts. Accordin gly, it is rem arkable th at
th e sh ieldin g effect of TBAP or TBATFB can be totally can celled by addition
of 0.1 M h alide supportin g electrolyte, im m ediately causin g curren t to jum p
to h igh values.
Th ese fin din gs suggest th at tetraalkylam m on ium cation s are specifically
adsorbed on th e silver cath ode, th us form in g a lipoph ilic film with th e in -
h ibitin g effect proportion al to its th ickn ess (i.e., to th e alkyl ch ain len gth );
th e above film is, h owever, im m ediately perturbed by addition of h alide
ion s, wh ich h ave even stron ger specific affin ity for silver. In fact, th ough
n egatively ch arged, h alide an ion s are reported to specifically adsorb on sil-
ver in aqueous solution s at quite n egative poten tials (–1.1/–1.2 V vs
SCE)12,13. Now, alth ough n o equivalen t studies are available for aceton itrile
0.6
0.5
0.6
0.5
a
b
TEAl
TEABr
TMAI
TBAI
0.4
0.3
0.4
0.3
TEAP
THepABr
TMATFB
TMAP
TMACl
TEATFB
THepAl
0.2
0.1
0.2
0.1
TBATFB
0.0
0.0
TBAP
–0.1
–0.1
0.000
0.003
0.006
0.009
–3
0.000
0.003
0.006
1, mol dm
0.009
–3
1, mol dm
FIG. 5
Supportin g electrolyte effect on acetobrom oglucose reduction peak curren ts (n orm alised to
100 m V s–1) in aceton itrile on silver wire electrodes (surface 0.16 cm 2, electrodeposition cur-
ren t den sity 0.5 m A cm –2). a Halide supportin g electrolytes: iodides (––––), brom ides (– – –),
ch loride (·····); b supportin g electrolytes: perch lorates (––––), tetrafluoborates (– – –)
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

Reactivity of Halo Sugars
891
solven t, it is quite reason able (an d our results m ay be regarded as a stron g,
th ough in direct eviden ce of th is h ypoth esis) th at in th is h igh ly polar,
aprotic m edium , h alide ion s m ay adsorb on silver even at m ore n egative
poten tials, eith er alon e or as an ion pair.
Th us it is reason able to state th at (i) in th e case of supportin g electrolyte
an ion s h avin g n o specific adsorption on silver, such as tetrafluoroborates
an d perch lorates, th e tetraalkylam m on ium cation s are specifically adsorbed
on silver so th at th e in n er Helm h oltz plan e (IHP) is a locus of positive
ch arges; (ii) in th e case of supportin g electrolytes h avin g stron g specific in -
teraction with silver, such as brom ide an d iodide, th e latter are specifically
adsorbed on silver eith er alon e or as an ion pair, so th at th e total ch arge on
th e IHP sh ould be n egative or, in an y case, less positive th an un der (i). Th is
effect of reversed IHP ch arge, h eavily affectin g th e activation en ergy profile
at th e electrode/solution in terface14 could explain th e greater activation en -
ergy required by th e process in th e presen ce of brom ides an d iodides.
It is rem arkable th at th e curren ts obtain ed with tetraeth ylam m on ium
salts, bein g th e h igh est in each an ion series, are also sh owin g th e lowest de-
pen den ce on th e an ion , eith er specifically in teractin g with silver or n ot.
Apparen tly, tetraeth ylam m on ium h as a very low film -over action so th at
th e active surface is sim ilar to th at obtain ed in th e presen ce of h alide ion s.
Furth er eviden ce could be provided by double-layer capacity studies an d by
SERS or STM (scan n in g tun n elin g m icroscopy)/AFM (atom ic force m icros-
copy) experim en ts.
Preparative Electrolysis Experiments
Cath odic Reduction of Halo Sugars with out Acceptors
Acetobrom oglucose (1) was also ch osen as a m odel substrate for a system atic
program of preparative electroreduction s on electrodeposited silver cath -
odes to test th e effect of operatin g con dition s on th e product distribution .
Th e workin g con dition s ch osen are sum m arised in Table III togeth er with
th e relevan t results wh ich are visualised in th e cen tral box of Sch em e 1.
In all cases, except wh en workin g with TMAP, th e m ain product is a sta-
tistical m ixture of th ree possible C–C glycosyl dim ers (α,α-, α,β-, an d β,β-),
wh ose form ation h as been ascribed to a radical in term ediate2–5 an d to its
peculiarly h igh local con cen tration on th e silver cath ode (a sort of cage ef-
fect).
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

892
Rondinini et al.:
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

Reactivity of Halo Sugars
893
As clearly sh own by experim en ts 1 to 3, to obtain m axim um acetobrom o-
glucose con version in to glycosyl dim ers (up to 80%), it is n ecessary to pre-
ven t th e process from run n in g in to diffusion -con trolled con dition s. In fact,
th e typical byproducts usually obtain ed3,4 wh en aceto- brom oglucose is to-
tally con sum ed durin g th e electrolysis (run I), drop alm ost to n il wh en elec-
trolysis is eith er stopped h alf-way (run II) or acetobrom oglucose is
periodically added to en sure a con stan t con cen tration (run III). Substitu-
tion of TEAP with TEABr does n ot sign ifican tly alter th e yields of both
Ph1: R1, R2, R3 = H
Ph2: R1 = OH, R2, R3 = H
AcO
AcO
silver cathode,
ACN/TeaP
Ph3: R1, R3 = H, R2 = OH
Ph4: R1, R2 = H, R3 = OCH3
Ph5: R1, R2 = H, R3 = CH2OH
O
O
OAc
OAc
OAc
OAc
+ 1e, - Br
Br
OAc
OAc
Ph6: R1 =
, R2, R3 = H
1
CN
OH
HO
+
R3
+
R1
R2
12 (major)
12 to 17
12 to 17
AcO
OAc
O
O
+
+
R1
R2
AcO
OAc
OAc
AcO
AcO
O
O
OAc
O
O
R3
OAc
OAc
OAc
OAc
OAc
AcO
AcO
OAc
18 (25-30%)
OAc
OAc
19 (up to 15%)
O
CN
OAc
OAc
OAc
OAc
AcO
OR'
OH
O
OAc
O
OAc
OAc
O
13
14
AcO
AcO
OAc
20
O
OAc
OAc
OAc
OAc
OAc
OAc
AcO
O
OAc
R' = H, OAc or
OAc
AcO
OAc
15
16
OAc
OAc
AcO
O
O
OAc
OAc
O
17
OAc
OAc
OAc
21
OH
SCHEME 1
A survey of acetobrom oglucose reaction s on silver cath odes
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

894
Rondinini et al.:
dim ers an d tetraacetylglucose, as th e byproducts in run IV, n ot yet com -
pletely elucidated, are very likely glucose-like com poun ds, probably
an om eric C–O–C dim ers.
Wh ile th e dim ers vs tetraacetylglucose ratio can be im proved by con trol-
lin g th e water con ten t an d tem perature (≈27 °C or lower), th e presen ce, yet
to be con firm ed, of C–O–C disacch arides could be related to th e periodic
acetobrom oglucose m ake-up.
Th e sim ilar product distribution in run s 3 an d 4 was n ot un expected,
sin ce brom ide ion s are always presen t in th e reaction m edium , as th ey form
durin g th e process. Th e cation effect, as poin ted out in th e previous sec-
tion , is m ore strikin g. In fact, th e lon ger-ch ain THepA ion gives alm ost on ly
dim ers an d tetraacetylglucose in th e 1 : 1 ratio (run V), wh ereas, rem arkably
en ough , ch an gin g TEAP for saturated TMAP (run VI) alm ost com pletely in -
h ibit dim erisation , wh ile four m on om eric products are obtain ed in sign ifi-
can t yields.
Electroch em ical Syn th esis of C–C Glycosyl Dim ers
Th e above procedure m ay be regarded as a powerful way to th e syn th esis of
both h om ogen eous an d crossed C–C glycosyl dim ers.
Now, th ese com poun ds are looked upon with in creasin g in terest by bio-
ch em ists as th ey are structurally very sim ilar to n atural C–O–C glycosyl
dim ers (in cludin g fun dam en tal disacch arides such as m altose, lactose, an d
sacch arose) but can n ot be m etabolised via h ydrolysis by th e sam e en zym es
(in th e above case, glucosidase, galactosidase, an d in vertase, respectively).
On th e oth er h an d, C–C glycosides are h ardly obtain able by n on -
electroch em ical path ways, as already m en tion ed above.
Th us we h ave carried out an explorative program of preparative
dim erisation s4 in cludin g differen t cases:
1. “Homo” dimerisations in the anomeric position, a case in wh ich th ree
products (α,α-, α,β-, an d β,β-) can be expected. After acetobrom oglucose, af-
fordin g h igh , statistical dim erisation yields (see th e previous section ), we
tested tri-O-acetyl-α-D-fucopyran osyl brom ide 6a h avin g a m eth yl group in -
stead of th e CH2OAc at th e C-6 position , obtain in g again all of th e th ree
possible isom ers th ough in lower yields (total 34%, wh ich , h owever, could
be im proved by optim izin g th e operation con dition s).
2. “Homo” dimerisation in non-anomeric, non-prochiral position, in wh ich
case on ly on e dim erisation product can be expected. Th e substrate tested,
6-deoxy-6-iodo-1,2:3,4-di-O-isopropyliden e-α-D-galactopyran ose 7, gave its
dim er as th e un ique product. Th is is a rem arkable result as th e in term ediate
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

Reactivity of Halo Sugars
895
radical is usually such a good h ydrogen abstractor from aceton itrile th at in
n on -electroch em ical experim en ts som e couplin g of th is radical was ob-
served on ly in th e presen ce of Fe(II); th is was in terpreted15 by th e h ypoth e-
sis of th e existen ce of a m etal ion /free radical com plex of a relatively lon ger
life th an a sim ple alkyl radical.
3. Cross-dimerisations: cogen eration by electroreduction on silver of a
n on -an om eric, n on -proch iral radical an d a m ore stable11 an om eric radical
(7 + 1; 7 + 6a). A partial selectivity was observed in both cases. Alth ough six
dim ers could be expected, products in total yields ≈40–60% alm ost en tirely
con sisted of th e th ree dim ers con tain in g th e m ore reactive radical 7 (ref.4).
From th ese prelim in ary results, glycosyl h alide electroreduction un doubt-
edly em erges as an extrem ely com petitive m eth odology for th e syn th esis of
C–C dim ers, but th e m eth od m ay turn out to be even m uch m ore efficien t
in term s of both dim erisation yields an d selectivity. In fact, all experim en ts
cited above were perform ed n ot on electrodeposited silver surfaces (i.e.,
h igh ly con trolled surfaces h avin g h igh rough n ess factor) but on silver
plates m ech an ically polish ed with san dpaper, an d, above all, th ey were car-
ried out on ly in stan dard “exh austive electrolysis” con dition s, as in run I,
Table III. Th ese con dition s proved by n o m ean s th e m ost con ven ien t for
dim erisation in th e recen t experim en ts on m odel acetobrom oglucose. Th us
work is in progress to apply th e im proved m eth odology to h om o- an d
cross-dim erisation s, an d to exten d it to polysacch aride dim erisation .
Electroch em ical Syn th esis of Glycosides
Reduction of acetobrom oglucose in th e presen ce of acceptors (Sch em e 1,
left an d righ t boxes) leads to th e form ation , som etim es stereoselective, of
glycosides, a widespread an d h igh ly valuable class of n atural com poun ds.
Reduction of acetobrom oglucose in th e presen ce of α,β-un saturated
n itriles led to th e stereospecific syn th esis of 3-glycosylalkan on itriles1. In
acrylon itrile glucosidation , th e yields were 26% on silver an d 19% on m er-
cury cath odes.
Reduction of acetobrom oglucose in th e presen ce of various ph en ols2,3 led
to th e form ation of th e correspon din g aryl O-glucosides in low to m oderate
yields. In th e case of 4-h ydroxyben zyl alcoh ol both th e aliph atic an d aro-
m atic glucoside were obtain ed. In th e m odel case of resorcin ol, six products
were isolated5, depen din g on operation con dition s, an α- an d a β-series,
both con sistin g of m on oglucosidated, glucosidated-acetylated, an d doubly
glucosidated resorcin ol.
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

896
Rondinini et al.:
A rem arkable effect of th e cath ode m aterial5 on th e product distribution
was observed. Workin g un der con trolled-poten tial con dition s on silver
cath ode, we obtain ed selectively α-glucosides an d C–C glucosyl dim ers as
m ain products (wh ile workin g un der backgroun d curren t con dition s we
also obtain ed β-glucosides, wh ich could be explain ed in term s of a SN2
m ech an ism started by th e ph en oxide an ion ); on palladium cath ode, both
α- an d β-glucosides an d again C–C glucosyl dim ers as m ain products; on
m ercury cath ode, α-glucosides in very low yields, an d tri-O-acetyl-D-glucal
as th e m ain product (in agreem en t with form er literature data16), wh ereas
th e yield of C–C dim ers dram atically decreased.
Th e observation th at con dition s prom otin g dim erisation s also prom ote
couplin g with oth er acceptors wh ereas con dition s prom otin g in tra-
m olecular reaction s con stan tly give low couplin g is con sisten t with our h y-
poth esis of th e “cage effect” provided by an active silver surface. Th e
glycosyl radicals, eith er true or laten t (th e brom ide actin g as a C···Br···Ag
lin k with th e silver surface), presen t in th e “cage” in h igh local con cen tra-
tion , couple with th eir n earest n eigh bours, eith er glycosyl radicals or accep-
tors. Th us dim erisation an d addition are n ecessarily always in com petition ,
th eir relative yields depen din g very likely n ot on ly on th e acceptor reactiv-
ity toward th e glycosyl radical, but also on th e relevan ce of its presen ce in -
side th e “cage”. In th is con text, th e presen t observation s about in teraction s
between th e silver surface an d adsorbed species could be very h elpful in de-
sign in g appropriate reaction con dition s to m axim ise th e couplin g yields.
Th e form ation of α-glycosides was explain ed by th e Giese–Dupuis h y-
poth esis17 th at th e glycosyl in term ediate partially m ain tain s sp3 h ybridisa-
tion , th e α con figuration bein g m ore stable th an th e β on e due to
in teraction between th e electron in an om eric position an d th e n on -
bon din g π orbital of th e adjacen t oxygen atom .
Som e exploratory cyclic voltam m etry experim en ts with h eterocyclic ac-
ceptors such as ben zim idazole an d purin e were also perform ed in order to
test th e possibility of C–N glycosidation , with n ucleosides as m ost valuable
targets; th ey gave eviden ce of specific in teraction between th e h alo sugar,
th e acceptor an d/or th e cath ode surface.
For preparative purposes, h owever, th e m eth odology so far used m ust be
radically ch an ged due to a very low solubility of h eterocyclic acceptors in
aceton itrile an d un suitability of th e custom ary direct-ph ase separation pro-
cedures.
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

Reactivity of Halo Sugars
897
CONCLUSIONS
Silver cath odes, h avin g un ique properties, can be regarded as a valuable
tool for glycosyl h alide reduction . Th ey actually sh ow th e best electro-
catalytic activity of th e tested m aterials. Th eir activity can be m odulated by
an appropriate ch oice of th e operatin g con dition s, in particular th e sup-
portin g electrolyte. Most im portan t of all, th ey exh ibit a rem arkable “cage
effect”, prom otin g h igh local con cen tration s of glycosyl radicals wh ich can
couple or add to oth er specifically adsorbed species.
Work is in progress to better en ligh t specific in teraction s between th e se-
lected substrates an d th e silver surface by m ean s of SERS (surface-en h an ced
Ram an spectroscopy), a powerful tech n ique for silver surface ch aracterisa-
tion .
To gen eralise th e peculiar beh aviour of organ ic h alides on silver, th e ex-
ten sion of th e presen t electroch em ical studies to sim ple alkyl h alides is n ow
n eeded. Actually th e reactivity of alkyl h alides h as been elucidated by very
careful an d th orough studies18, yet gen erally referrin g to cath odes (plati-
n um , glassy carbon ) h avin g little specific in teraction with th e selected sub-
strates. In th is con text, testin g silver could lead to detection of n ew an d
rem arkable features.
The authors are grateful to Prof. S. Trasatti for helpful discussion and to the National Research
Council of Italy (C.N.R.) for financial support.
REFERENCES
1. Rondinini S., Mussini P. R., Ferzetti V., Monti D.: Electrochim. Acta 1991, 36, 1095.
2. Benedetto M., Miglierini G., Mussini P. R., Pelizzoni F., Rondinini S., Sello G.: Carbohydr.
Lett. 1995, 1, 321.
3. Rondinini S., Mussini P. R., Sello G., Vismara E.: J. Electrochem. Soc. 1998, 145, 1108.
4. Guerrini M., Mussini P., Rondinini S., Torri G., Vismara E.: Chem. Commun. 1998, 1575.
5. Rondinini S., Mussini P. R., Cantu G., Sello G.: Phys. Chem. Chem. Phys. 1999, 1, 2989.
6. Alberti A., Della Bona M. A., Macciantelli D., Pelizzoni F., Sello G., Torri G., Vismara E.:
Tetrahedron 1996, 52, 10241.
7. Ives D. J. G., Janz J. G.: Reference Electrodes – Theory and Practice, p. 203. Academic Press,
New York 1961.
8. Trasatti S., Petrii O. A.: Pure Appl. Chem. 1991, 63, 711.
9. Kolb D. M., Przanyski M., Gerischer H.: J. Electroanal. Chem. Interfacial Electrochem. 1974,
54, 25.
10. a) Vashkyalis A., Demontaite O.: Elektrokhimiya 1978, 14, 1213; b) Vashkyalis A.,
Demontaite O.: Sov. Electrochem. 1979, 1950.
11. Comoli M.: Ph.D. Thesis. University of Milano, Milano 1998.
Collect. Czech. Chem. Commun. (Vol. 65) (2000)

898
Rondinini et al.:
12. a) Valette G., Hamelin A., Parsons R.: Z. Phys. Chem. 1978, 113, 71; b) Valette G.:
J. Electroanal. Chem. 1982, 132, 311; c) Valette G.: J. Electroanal. Chem. Interfacial
Electrochem. 1983, 146, 439; d) Valette G., Parsons R.: J. Electroanal. Chem. Interfacial
Electrochem. 1985, 191, 377; e) Valette G., Parsons R.: J. Electroanal. Chem. Interfacial
Electrochem. 1986, 204, 291; f) Bange K., Straehler B., Sass J. K., Parsons R.: J. Electroanal.
Chem. Interfacial Electrochem. 1987, 229, 87.
13. a) Foresti M. L., Innocenti M., Kobayashi H., Pezzatini G., Guidelli R.: J. Chem. Soc.,
Faraday Trans. 1996, 92, 3747, and references therein; b) Foresti M. L., Innocenti M.,
Forni F., Guidelli R.: Langmuir 1998, 14, 7008, and references therein.
14. a) Conway B. E.: Theory and Principles of Electrode Processes, Chaps 4 and 5. Ronald, New
York 1965; b) Bianchi G. in: Elettrochimica (T. Mussini, Ed.), p. 476. Noguerol, Milano
1963
15. Kochi J., Rust F. F.: J. Am. Chem. Soc. 1962, 83, 2017.
16. Giese B., Dupuis J.: Tetrahedron Lett. 1984, 25, 1349.
17. Maran F., Vianello E., Catellani G., D’Angeli F.: Electrochim. Acta 1989, 34, 587.
18. a) Andrieux C. P., Blocman C., Dumas-Bouchiat J.-M., Savéant J.-M.: J. Am. Chem. Soc.
1979, 101, 3431; b) Lexa D., Savéant J.-M., Soufflet J. P.: J. Electroanal. Chem. 1979, 100,
159; c) Andrieux C. P., Merz A., Savéant J.-M., Tomahogh R.: J. Am. Chem. Soc. 1984,
106, 1957; d) Andrieux C. P., Merz A., Savéant J.-M.: J. Am. Chem. Soc. 1985, 107, 6097;
e) Andrieux C. P., Gallardo I., Savéant J.-M., Su K. B.: J. Am. Chem. Soc. 1986, 108, 638;
f) Andrieux C. P., Savéant J.-M., Su K. B.: J. Phys. Chem. 1986, 90, 3815; g) Savéant J.-M.:
J. Am. Chem. Soc. 1987, 109, 6788; h) Chami Z., Gareil M., Pinson J., Savéant J.-M.,
Thiébault A.: Tetrahedron Lett. 1988, 29, 39; i) Andrieux C. P., Hapiot P., Savéant J.-M.:
J. Phys. Chem. 1988, 92, 5887; j) Lexa D., Savéant J.-M., SuK. B., Wang D. L.: J. Am.
Chem. Soc. 1988, 110, 7617; k) Andrieux C. P., Gallardo I., Savéant J.-M.: J. Am. Chem.
Soc. 1989, 111, 1620; l) Lexa D., Savéant J.-M., Schäfer H., SuK. B., Vering B.,
Wang D. L.: J. Am. Chem. Soc. 1990, 112, 6162; m) Chami Z., Gareil M., Pinson J.,
Savéant J.-M., Thiébault A.: J. Org Chem. 1991, 56, 586; n) Médebielle M., Pinson J.,
Savéant J.-M.: J. Am. Chem. Soc. 1991, 113, 6872; o) Andrieux C. P., Le Gorande A.,
Savéant J.-M.: J. Am. Chem. Soc. 1992, 114, 6892; p) Bertran J., Gallardo I., Moreno M.,
Savéant J.-M.: J. Am. Chem. Soc. 1992, 114, 9576; q) Savéant J.-M.: J. Am. Chem. Soc.
1992, 114, 10595; r) Ettayeb R., Savéant J.-M., Thiébault A.: J. Am. Chem. Soc. 1992, 114,
10990; s) Adcock W., Clark C., Houmam A., Krstic A. R., Pinson J., Savéant J.-M.,
Taylor D. K., Taylor J. F.: J. Am. Chem. Soc. 1994, 116, 4653; t) Andrieux C. P.,
Le Gorande A., Savéant J.-M.: J. Electroanal. Chem. 1994, 377, 191; u) Andrieux C. P.,
Batlle A., Espin M., Gallardo I., Jiang Z., Marquet J.: Tetrahedron 1994, 50, 6913;
v) Savéant J.-M.: Adv. Electron Transfer Chem. 1994, 4, 53; w) Andrieux C. P., Robert M.,
Savéant J.-M.: J. Am. Chem. Soc. 1995, 394, 9340; x) Médebielle M., Pinson J.,
Savéant J.-M.: Electrochim. Acta 1997, 42, 2049; y) Andrieux C. P., Saveant J.-M., Tardy C.:
J. Am. Chem. Soc. 1998, 120, 4167.
Collect. Czech. Chem. Commun. (Vol. 65) (2000)