Behaviour of xyloisosaccharinic acid and xyloisosaccharino-1,4-lactone in aqueous solutions at varying pHs

Article abstract of DOI:10.1016/j.carres.2012.09.019

Xyloisosaccharinic acid is one of the major degradation products formed during the alkali catalysed hydrolysis of hemicelluloses. In acidic solution xyloisosaccharinic acid undergoes an acid catalysed lactonisation to generate xyloisosaccharino-1,4-lacton

Full text of DOI:10.1016/j.carres.2012.09.019

Carbohydrate Research 363 (2012) 51–57
Contents lists available at SciVerse ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Behaviour of xyloisosaccharinic acid and xyloisosaccharino-1,4-lactone
in aqueous solutions at varying pHs
Michael Almond a, Paul B. Shaw a, Paul N. Humphreys a, Marcus J. Chadha b, Klaus Niemelä c,
Andrew P. Laws a,
⇑
a Department of Chemical and Biological Sciences, School of Applied Sciences, University of Huddersfield, Queensgate, Huddersfield HD1 3DH, United Kingdom
b IPOS, University of Huddersfield, Queensgate, Huddersfield HD1 3DH, United Kingdom
c VTT Technical Research Centre of Finland, Tietotie 2, Espoo, PO Box 1000, FI-02044 VTT, Finland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 August 2012
Received in revised form 24 September 2012
Accepted 26 September 2012
Available online 11 October 2012
Xyloisosaccharinic acid is one of the major degradation products formed during the alkali catalysed
hydrolysis of hemicelluloses. In acidic solution xyloisosaccharinic acid undergoes an acid catalysed lac-
tonisation to generate xyloisosaccharino-1,4-lactone. We report here the solution phase properties of
xyloisosaccharinic including measurement of its aqueous pKa (3.00 0.05) using 13C NMR methods.
We also report rate constants for the acid catalysed lactonisation, klact(D20), of xyloisosaccharinic acid
and the results of our investigations of the kinetics of hydrolysis of xyloisosaccharino-1,4-lactone at
Keywords:
Xyloisosaccharinic acid
2,4-Dihydroxy-2-hydroxymethylbutanoic
acid
acidic and basic pHs. The second-order rate constants for the hydrolysis reactions kHO– (25 Mꢀ1 sꢀ1
)
and kD+ (4.13 E-4 Mꢀ1 sꢀ1).
Ó 2012 Elsevier Ltd. All rights reserved.
3-Deoxy-2-C-hydroxymethyl-
acid
D,L-tetronic
Xyloisosaccharino-1,4-lactone
pKa measurement
1. Introduction
properties of many important saccharinic acids. Whilst a number
of workers have explored the properties of isomeric glucoisosac-
charinic acids (GISAH)21–30 very little work has been undertaken
looking at the properties of xyloisosaccharinic acid (XISAH,
Saccharinic acids are a group of polyhydroxylated monocarbox-
ylic acids and are the main decomposition products generated
when polysaccharides and other carbohydrate materials are trea-
ted with aqueous alkaline solution.1,2 They are formed in especially
high amounts in alkaline wood pulping processes.3 Saccharinic
acids are of interest because they are potentially useful raw mate-
rials; they are good metal chelating agents4–11 and, in their enan-
tiomerically pure form, they are valuable carbon skeletons with
predefined stereochemistry that can be easily functionalised for
use in synthesis.12–16 Saccharinic acids also have the potential to
leach metal ions from cellulosic waste materials; this is of particu-
lar concern for the design of the deep-underground repositories
currently under consideration17–20 for the disposal of intermediate
level radioactive waste. It is suspected that the production of sac-
charinic acids in the environment of a nuclear waste repository
could influence the containment of radionuclides. Despite their
importance, there are still gaps in our knowledge of the physical
3-deoxy-2-C-hydroxymethyl-D,L-tetronic acid). Whistler and
Corbett31 and Aspinall et al.32 simultaneously demonstrated that
XISAH is produced when small xylooligosaccharides are reacted
with aqueous alkali. It has also been shown that significant
amounts of XISAH are generated during alkaline treatment of
corn33,34 and birch33,35 xylans. The alkali catalysed degradation of
xylans is also responsible for the dominant production of XISAH
during the alkaline pulping of hardwoods.35,36 Like the corre-
sponding GISAH, in acidic solution XISAH lactonises to give the
corresponding xyloisosaccharino-1,4-lactone (XISAL, 3-deoxy-2-C-
hydroxymethyl-D,L-tetrono-1,4-lactone, Scheme 1). Alén and Val-
konen have crystallised XISAL and published its X-ray structure.37
Despite its potential value, little is known about the solution
phase physical properties of XISAH. We report here the solution
phase properties of XISAH including measurement of the aqueous
pKa of XISAH using 13C NMR methods. We also report rate con-
stants for the acid catalysed lactonisation of XISAH and the results
of our investigations of the kinetics of hydrolysis of XISAL at both
acidic and basic pHs.
⇑
Corresponding author. Tel.: +44 (0) 1484 472668; fax: +44 (0) 1484 472182.
E-mail address: a.p.laws@hud.ac.uk (A.P. Laws).
0008-6215/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.carres.2012.09.019

52
M. Almond et al. / Carbohydrate Research 363 (2012) 51–57
O
HO
O
HO
HOH2C
HO
HOH2C
O
HO
HOH2C
kHYDROLYSIS
Ka
O
O
OH
kLACTONISATION
OH
XISAH
Scheme 1.
XISAL
XISA
round bottom flask and the solution was stirred under an atmo-
sphere of nitrogen, The reaction vessel was attached to a Metrohm
autotitrator (Titrano model 857, Metrohm, Runcorn, UK) and was
maintained at 25 °C throughout the experiments. Fresh sodium
hydroxide solutions were used as the titrant, these were prepared
from a volumetric standard prior to the reaction and the titrant
was flushed with nitrogen to avoid absorption of carbon dioxide.
The appropriate start pH was selected and the volume of titrant
added as a function of time was recorded. All pH measurements
2. Experimental
2.1. Materials and chemicals
Xyloisosaccharino-1,4-lactone (XISAL, 3-deoxy-2-C-hydroxy-
methyl- -tetrono-1,4-lactone) was prepared by GalChimia (Santi-
D,L
ago de Compostela, Spain) using the procedures described by
Aspinall et al.32 The purity of the XISAH was determined by conver-
sion to its pertrimethylsilyl-derivative and analysis by GC–MS. The
product was trimethylsilylated with a mixture (1:1) of BSTFA and
TMCS and was analysed with an Agilent 6890 Series GC System,
equipped with an Agilent 5973 Mass Selective Detector and a Phe-
nomenex ZB-5HT Inferno capillary column (30 m ꢁ 0.25 mm, film
were recorded in-situ using
a Unitrode combined glass pH
electrode equipped with a built-in PT1000 temperature sensor
(Metrohm, Runcorn, UK).
2.4. Following the inter-conversion of XISAH and XISAL in acidic
solution
0.25 lm). The temperature programme was 3 min at 200 °C, fol-
lowed by 8 °C/min to 320 °C (for 15 min). The results indicated c
99% purity of XISAL, and the mass spectrum was in good agree-
ment with the literature data.38 MS (m/z, rel intensity): 261 [10
(M-15)], 246 (100), 233 (8), 190 (10), 147 (77), 143 (35), 103
(27), 73 (77).
The kinetics for the lactonisation of XISAH were followed using
NMR spectroscopy. Samples of XISAL were dissolved in D2O
(20 mg/mL) and the pH of the solution was initially increased to
12 and the sample was left for ten minutes to ensure complete
conversion of the XISAL to the anion XISA. The sample was cooled
in an ice bath before adjusting the pH of the solution to the re-
quired pD (1–2.5) through the addition of DNO3. Samples were
immediately transferred to the NMR spectrometer and spectra
were recorded at regular periods over a period of several days or
until no further reaction was observed. When spectra were not
being recorded samples were stored at 25 °C; the pD of the reac-
tion solution was monitored throughout the course of the reaction.
An attempt was made to monitor the equilibrium between the
lactone and the free acid using NMR spectroscopy: samples of the
lactone were dissolved in D2O (10 mg/mL) and the pD of the solu-
tion was reduced to acidic pDs through the addition of DNO3. A ser-
ies of solutions of varying pD (1–5) were prepared and these were
stored, under an atmosphere of nitrogen, at 25 °C for a period of up
to two months.
All other reagents were analytical grade reagents and were pur-
chased from the Aldrich Chemical Company (Poole, UK).
2.2. Determination of the aqueous pKa of XISAH
XISAL (20 mg, 0.151 mmol) was dissolved in D2O (1 mL) and the
pH of the solution was varied between pD 1 and 11 using either
DNO3 or NaOD and reporting pD as the measured pH +0.4. Samples
were quickly transferred to NMR tubes and spectra were recorded
at a probe temperature of 300 K on a Bruker Avance 500 MHz
(125 MHz 13C) spectrometer. Proton decoupled and NOE enhanced
carbon spectra were recorded using 256 scans employing a wait
time between scans of 2 s for experiments where chemical shifts
were being measured and 10 s for those spectra where carbon sig-
nals were being integrated; ten seconds was chosen to allow com-
plete relaxation (>5 ꢁ T1) of the carbon resonances used in the
kinetic studies. A capillary insert, containing CDCl3, was used as
an internal reference. For the assignment of resonances to individ-
ual carbons in XISAH and XISAL a series of 1D and 2D-NMR spectra
were recorded at both high and low pHs and samples were allowed
to fully equilibrate before spectra were recorded, that is, until sin-
gle sets of resonances were observed, at extremes of pH, less than
30 mins was needed for complete transformation to either XISAL or
XISA. Spectra recorded included proton, Carbon DEPT, COSY, HMBC
and HSQC and these were acquired using Bruker standard pulse
sequences. The solution pH was measured in the NMR tube using
a Beckman ø40 pH Meter (High Wycombe, UK) in combination
with a Hanna instruments glass long reach NMR pH electrode
(Mannheim, Germany). The pKa of the acid was determined using
the Henderson–Hasselbach equation: pKa = pD +log{[AD]/[Aꢀ]}.39,40
2.5. Analysis of the mechanism of the base catalysed sovolysis
reactions
For the hydrolysis reactions using oxygen-18 labelled water, XI-
SAL (10 mg) was dissolved in a solution of 18O-labelled H2O (98%
1 mL) and the pH of the solution was maintained at 7 for ten days
(0.1 M NaOH was added as required). After ten days the sample
was diluted with ultra-pure water before analysis by LC–MS–MS
using an Agilent 6460 triple quadrupole mass spectrometer
(Cheadle, UK) fitted with a JetStream electrospray source interfaced
with an Agilent 1290 HPLC. Samples (5
ll) were separated on a
Zorbax Extend C-18 column (1.8
lm particle size and 2.1 ꢁ 50 mm)
using a mobile phase containing acetonitrile (A) and aqueous
ammonium acetate (20 mM, B) operating with isocratic elution
(90% A:10% B) with a flow rate of 0.25 mL/min and at a column tem-
perature of 25 °C. The electrospray interface was operated using
negative ion polarity. The JetStream source gas temperature was
maintained at 300 °C, with a gas flow of 10 L/min and a nebuliser
gas pressure of 15 psi. An optimal spray was obtained with the
capillary voltage and the nozzle voltage set at 3500 and 500 volts,
2.3. Following the kinetics of the hydrolysis of XISAL in alkaline
solution
A stock solution of XISAL (20 mg/100
pared prior to experiments. Kinetic runs were initiated by adding
the XISAL (50 L) stock solution to ultrapure water (65 mL) in a
lL HCl (0.1 M)) was pre-
l

M. Almond et al. / Carbohydrate Research 363 (2012) 51–57
53
respectively, and employing a delta EMV of 150 volts. The sheaf gas
temperature was maintained at 250 °C and a flow rate of 7 L/min.
The mass spectrometer was operated using product ion scanning,
for the parent ion the fragmentation voltage was set at 100 volts
and collision-induced dissociation was monitored over a range of
collision energies between 5 and 20 volts. For the pseudo-MS–MS
the fragmentation voltage was increased to 150 volts in order to
favour production of the required daughter ions.
signed to the quaternary centres of XISA, that is, C1 & C2, the
lowest field signal being assigned to the carbonyl carbon
(C1,179.9 ppm) and the second signal to C2 (78.3 ppm). Analysis
of a combination of the COSY and HSQC spectra allowed the assign-
ment of the remaining two methylene groups: the C4 methylene
gives rise to a signal at 57.8 whilst the C5 methylene gives rise
to the resonance centred at 66.6 ppm.
When a solution of XISAH was acidified to a pH less than 3, the
initial spectrum contains the same five resonances, although some
have moved significantly (see discussion below) and this is consis-
tent with protonation of the anion and formation of the free acid
XISAH in solution. However, a further five peaks slowly appeared
over time and these additional peaks were attributed to being
those of the lactone (XISAL, top trace Fig. 1). A similar set of
DEPT-135 and 2D-COSY, HMBC and HSQC spectra were used to
confirm the location of the individual resonance lines of the car-
bons in the lactone: C1 (180.4 ppm) C2 (75.9 ppm) C3 (32.7 ppm)
C4 (67.2 ppm) and C5 (64.3 ppm). The down-field shift of C4
matches a similar shift observed for C4 of alpha-GISA25 on lacton-
isation and is consistent with acylation of the primary alcohol and
formation of a five member-ring lactone.
3. Results and discussion
Xyloisosaccharinic acid (XISAH) is a small polyhydroxylated ali-
phatic carboxylic acid and, in aqueous solution, will exist as the
free acid (XISAH) at moderately acidic pHs and as its conjugate
base (XISA) at pHs significantly above the acid’s pKa. In common
with other 4-hydroxy-substituted monocarboxylic acids, XISAH is
known to cyclise at low pHs to form xyloisosaccharino-1,4-lactone
(XISAL).32 In order to determine the relative proportions of the
three species present, a study of the variation of the position and
number of the 13C NMR resonances of a solution of XISAH was
undertaken at various pHs.
At neutral and alkaline pHs the 13C spectrum recorded for XI-
SAH contains five resonances corresponding to the five carbon
atoms of the anion (XISA, bottom trace Fig. 1). The identity of the
individual signals was determined by analysis of both the carbon
spectrum and the carbon DEPT-135 spectra (not shown): the three
methylene groups gave negative signals on the DEPT-135 spectrum
and the signal at lowest field was assigned to C3 (36.7 ppm) the
other two methylene resonances are shifted to higher field by
the presence of the electron withdrawing hydroxyl groups. The
two carbon signals missing from the DEPT-135 spectrum were as-
In order to determine the aqueous pKa of the carboxylic acid of
XISAH, the variation in the chemical shifts of the carbons signals of
XISA/H with the pD of the solution was monitored. The chemical
shift for three of the five carbons varied and a sigmoidal depen-
dence of the chemical shift of C1, C2 and C4 on the pD of the solu-
tion was observed (Fig 2a–c).
As was expected, the largest shift was observed for the carbonyl
carbon (C1
Dd = 2.22 ppm) identifying that the sigmoidal depen-
dence is related to protonation of the carboxylate anion. The up-
field shift of the carbonyl carbon signal is similar to that observed
C2-L
C4-L C5-L
C3-L
pH=0.28
pH=0.28
pH=3.04
pH=3.04
pH= 5.02
pH= 5.02
pH= 6.00
pH= 6.00
CHCl3
C2
C5
C3
C4
190
185
180
175
170
80
75
70
65
60
55
50
45
40
35
ppm
ppm
Figure 1. 13C chemical shifts (d in ppm) for the resonances of XISA/H (for assignment of the signals in XISA see the bottom spectrum) and XISAL with residual XISAH (for
assignment of the carbon signals in the lactone-see the top spectrum) at various pDs recorded at 300 K.

54
M. Almond et al. / Carbohydrate Research 363 (2012) 51–57
for protonation of the carboxylate group in other saccharinic
acids.25,41 As proton transfer between the free acid and anion will
be rapid, the observed chemical shift d(pD) for each carbon signal
will reflect the mole fraction of each species present and can be fit-
ted to the equation:
followed over time. At a fixed pH, the volume of base consumed
in the hydrolysis reactions decreased exponentially with time indi-
cating a first-order reaction and a plot of the logarithm of the vol-
ume of base added against time was linear from which it was
possible to calculate a pseudo-first order rate constant (kobs) for
the hydrolysis at that pH. This process was repeated for a range
of hydroxide concentrations and a pH rate profile for the base cat-
alysed hydrolysis reaction was constructed (Fig 3). From inspection
of the pH rate profile it is clear that at the lowest pHs studied the
hydrolysis of the lactone is extremely slow (t1/2 pH 7 of 26 h)
whilst at high pHs the hydrolysis reaction is very fast (t1/2 pH
10.5 of 90 s). The reactions between pH 8.5 and 10.5 were allowed
to run to completion (5 ꢁ t1/2) during which period one equivalent
of base was consumed indicating that above the pKa of the acid the
lactone (XISAL) does not exist in equilibrium with the anion (XISA)
and that if a solution of the lactone is left for long-enough it will
undergo complete hydrolysis. At high pHs the rate profile has a
slope of one and this is consistent with a bimolecular base cata-
lysed reaction. The vast majority of simple esters undergo base cat-
alysed hydrolysis via a BAC2 mechanism, that is, through nucle-
ophilic attack at the carbonyl carbon, however, at neutral pHs a
number of substituted cyclic lactones including b-butyrolactone44
are able to react via attack of the nucleophile at the alkoxy-carbon
with accompanying cleavage of the alkoxy-oxygen bond, that is,
via a BAL2 mechanism. The BAL2 mechanism normally only operates
when attack of the nucleophile at the acyl-carbon is blocked by
bulky substituents located adjacent to the carbonyl group. In the
X-ray structure there is evidence for crowding of the carbonyl
centre by the adjacent C2 hydroxyl and hydroxymethylene
substituents. In order to get information about the position of
attack of the nucleophile in XISAL, the hydrolysis reaction was per-
formed at pH 7 in 18O-labelled water and the location of the label
in the product was identified by LC–MS–MS. Mass spectra obtained
for the reactions performed in 16O and 18O labelled water (Fig 4)
show the expected molecular ions for XISA at m/z of 151 Da for
the 18O labelled nucleophile and at m/z of 149 for the 16O labelled
nucleophile. The highest mass fragment in both systems arises
from the expulsion of the C2-hydroxymethylene group, as
formaldehyde, via a McLafferty-type rearrangement (Scheme 2).
A similar rearrangement has been reported for 2,3-dihydroxyprop-
anoic acid which possesses the same b-hydroxycarbonyl group
that is present in XISA.45 The McLafferty rearrangement initially
generates a dihydroxyenolate which will tautomerise to form the
2,4-dihyroxybutanoate anion which is visible at m/z = 121(18O)
dðpDÞ ¼ dacid þ dbase½10ꢀpD=ðKa þ 10ꢀpDÞꢂ
The measured pKa’s from the individual resonances are 3.37
(C1) 3.47 (C2) and 3.36 (C4) with the mean value being
3.40 0.05. Glasoe and Long42 established a
D
pKa value of ꢀ0.4
for transfer from D2O to H2O for a range of weak acids and there-
fore the corresponding pKa in H2O can be estimated as 3.0. The
measured pKa is slightly lower than that measured for the two dia-
stereoisomers of GISAH recorded using NMR chemical shifts (al-
pha-GISAH = 3.3625 and beta-GISAH = 3.6141) and is lower than
the value of 3.49 calculated by Käkölä et al. for XISAH.43 Given that
the substituents immediately adjacent to the carboxylate group are
the same in XISAH and GISAH it is difficult to account for the in-
creased acidity of XISAH. However, there is some evidence for
the presence of a hydrogen-bonding interaction between the car-
boxylate group and the hydroxyl substituent at C4. The sigmoidal
dependence of the C4-chemical shift of XISAH as a function of pD
and the significant movement in the resonance position of the C4
carbon on protonation of the carboxylate group suggests that there
is direct interaction between the substituents at these two centres.
Cho et al.25 suggested that for alpha-GISAH a hydrogen bond exists
between the carboxylate group and the secondary alcohol at C4, a
similar hydrogen bond in XISAH could account for the movement
in the chemical shift for C4 observed in the present study. Differ-
ences in the extent of this hydrogen-bonding, which involves a pri-
mary hydroxyl group in XISAH and a secondary hydroxyl group in
GISAH, could potentially account for the observed difference in the
measured pKa values.
Our NMR studies indicated that at low pHs XISAH lactonises to
form XISAL and we were interested in measuring the rate con-
stants for the inter-conversion of the lactone (XISAL) and the free
acid (XISAH). When XISAL was dissolved in D2O at neutral pH’s
(6–8) the NMR spectrum that was initially recorded included large
signals for the lactone, the intensity of these signals slowly reduced
over-time and the drop in signal intensity was accompanied by a
corresponding drop in the pH of the sample, indicating that lactone
hydrolysis was occurring. In order to monitor the kinetics of the
hydrolysis reaction, samples of XISAL were hydrolysed at fixed
pHs (7–10.5) in an autotitrator and the consumption of base was
180.0
179.5
179.0
178.5
178.0
177.5
78.2
78.1
78.0
77.9
77.8
77.7
77.6
77.5
77.4
77.3
a
b
c
57.7
57.6
57.5
57.4
57.3
57.2
57.1
57.0
56.9
0
1
2
3
4
5
6
7
0
1
2
3
4
5
6
7
0
1
2
3
4
5
6
7
pD
pD
pD
Figure 2. Variation of the 13C chemical shifts of XISAH as a function of pD. Solid line represents the fit to the equation: d(pD) = dacid + dbase[10–pD/(Ka + 10–pD)] giving a mean
pKa in D2O of 3.40 0.05.

M. Almond et al. / Carbohydrate Research 363 (2012) 51–57
55
a
b
0.008
-2.0
-2.5
-3.0
-3.5
-4.0
-4.5
-5.0
-5.5
0.006
0.004
0.002
0.000
-0.00005 0.00000 0.00005 0.00010 0.00015 0.00020 0.00025 0.00030 0.00035
6.5
7.0
7.5
8.0
8.5
9.0
9.5 10.0 10.5 11.0
[HO-]/M
pH
Figure 3. Left-hand side: pH versus rate profile for the hydrolysis of XISAL at 25 °C. Right-hand side: plot of kobs versus [HOꢀ] for the hydroxide catalysed hydrolysis of XISAL.
Figure 4. Top, mass spectrum for an XISA sample prepared by reaction of XISAL with H2 18O/NaOD which was maintained at pH 7 and at room temperature for 12 days and
employing a collision energy of 10 volts. Bottom, mass spectrum for an XISA sample prepared by reaction of XISAL with H2 16O/NaOD which was maintained at pH 7 and at
room temperature for 12 days and employing a collision energy of 10 volts. Further details of how the spectra were recorded are listed in the Section 2.
and 119 (16O). The next highest mass fragments, m/z = 103 (18O)
and 101 (16O), can be attributed to the loss of water from 2,4-
dihydroxybutanoic acid and this was confirmed by studying the
daughter ions generated by collision-induced dissociation (CID)
of the m/z 121 and 119 ions. The third highest mass fragment is
the most abundant fragment and in both systems this fragment oc-
curs at m/z of 73, signifying that the isotopic label has been lost as
part of a neutral molecule, this is likely to arise from the elimina-
tion of the elements of formic acid in the concomitant loss of car-
bon dioxide and molecular hydrogen from 2,4-dihyroxybutanoate.
That the m/z of 73 is generated from both the 119 and 121 ions was
confirmed when it was produced as a daughter during CID of the
m/z 121 and 119 ions. A very similar fragmentation has previously
been reported by Bowie and co-workers46 in the spectra obtained

56
M. Almond et al. / Carbohydrate Research 363 (2012) 51–57
H
O
H
O 16/18
16/18
O
16/18
O
O
OH
H
OH
H
OH
OH
H
O
OH
O
HO
HO
HO
H
H2CO
H
H
H
H
OH
H2
+
CO2
HO
m/z =151/149
m/z =73
m/z =121/119
m/z =121/119
OH
OH
16/18
O
OH
HO
OH
O
O
16/18
O
+
O
H
H
H
OH
OH
OH
m/z =151/149
m/z =45/47
Scheme 2.
for CID of carboxylate ions that are not able to form stabilised car-
banions. The loss of CO2 and H2, and the accompanying loss of mass
of 48 and 46 in the two systems, clearly identify that the label is
present in the carboxylate group and this infers that lactone hydro-
lyses occur via the BAC2 mechanism. A second piece of evidence
that supports the location of the label in the carboxylate group is
the observation of fragments at m/z 47 (18O) and 45 (16O) as abun-
dant fragments of the molecular ion but only as minor ions in the
spectrum formed from the CID of m/z 121 and 119. The two frag-
ments are likely to be derived from the elimination of formate an-
ion from XISA (Scheme 2).
temperature of 25 °C. These experiments were followed by
monitoring the ratio of XISAH:XISAL using 13C NMR (Fig 5). The
XISAH concentration decayed exponentially with time allowing
the half-life for the equilibration reaction to be determined
(pH 1.1 t1/2 = 4.75 h and at pH 1.97 t1/2 = 48.2 h). Assuming both
the forward and reverse reactions are acid catalysed, this would
suggest that the half-life for the equilibration process would be
approximately 500 and 5000 h at pHs 3 and 4, respectively. Given
the slow rates of reaction, the time for complete equilibration at
pH 3, that is, 5 ꢁ t1/2, would be approximately three months and
this figure is consistent with the very slow conversions observed
in the first set of experiments.
A measure of the relative reactivity of XISAL can be obtained by
determining the second-order rate constant for hydrolysis with
those measured for other lactones. Using a value of 1.01 ꢁ 10ꢀ14
for the ion product of water, Kw, the second order rate constant
for the hydroxide catalysed hydrolysis of XISAL can be determined
from the slope of a plot of hydroxide concentration against kobs
(kobs = kHO–[HO–], Fig 3b). The value obtained kHO– = 25 Mꢀ1 sꢀ1 is
approximately 25-fold lower than the value estimated for
butyrolactone, kHO– = 1.0 Mꢀ1 sꢀ1, determined by extrapolation
of data reported by Blackburn and Dodds47 measured at 46 °C
(kHO– = 4.03 Mꢀ1 sꢀ1). One possible explanation for the increased
reactivity of XISAL is that the adjacent electron withdrawing hy-
droxyl group activates the carbonyl carbon to nucleophilic attack.
In acidic solution, XISAL can exist in equilibrium with XISAH.
Normally, this would offer the opportunity to study the equilibra-
tion process and allow us to determine the relative abundance of
the lactone and free acid forms present at different pHs, similar
studies have been performed on GISAH/L.48 In a first attempt to
monitor the relative proportions of the two species XISAH and XI-
SAL present, a number of experiments were performed in which a
solution of the XISAL was prepared in D2O at acidic pDs, between 0
and 5, and these were left in the dark under an atmosphere of
nitrogen for a period of up to two months and NMR spectra were
recorded periodically through-out the course of the reaction.
Unfortunately, at the higher pDs, 4 and 5, the time taken to reach
equilibrium was too long and the free acid was only present in very
small amounts. In order to overcome the small extent of reaction, a
second set of experiments was performed but starting in the re-
verse direction, that is, adding the XISAH to acidic solution and
monitoring the loss of XISAH and the production of XISAL at a fixed
For the reactions performed at pDs between 1 and 2, the ob-
served rate constant (kobs) measured for equilibration at a fixed
pD is the sum of the forward (klact/D2O)[D+] and reverse rate con-
stants (khyd/D2O)[D+] and given that the equilibrium concentration
(Kequ/D2O) is equal to the forward rate constant (klact/D2O) divided by
the reverse rate constant (khyd/D2O) and is equal to the relative pro-
portions of XISAH/L present at the end of the reaction, it is possible
110
pD 1.49
pD 2.37
100
90
80
70
60
50
40
30
20
-1000
0
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000
Time/min
Figure 5. Plot of the % XISAH remaining in acidic solution (pD 1–2.5) as a function
of time, the % values were calculated from the integration of the carbon signals
recorded at 300 K and using extended delays between pulses for quantification (see
Section 2).

M. Almond et al. / Carbohydrate Research 363 (2012) 51–57
3. Sjöström, E. Tappi 1977, 60, 151.
57
Table 1
4. Svensson, M.; Berg, M.; Ifwer, K.; Sjoblom, R.; Ecke, H. J. Hazard. Mater. 2007,
144, 477.
5. Gaona, X.; Montoya, V.; Colas, E.; Grive, M.; Duro, L. J. Contam. Hydrol. 2008,
102, 217.
Observed rate constants (kobs/s-1) and equilibrium constant (Kequ/D2O) measured for
the lactonisation of XISAH at various values for pD determined by following the
exponential decay of the XISAH as a function of time using NMR at 25 °C. The second-
order rate constants for the acid catalysed lactone hydrolysis (khyd(D2O)[D+]/s-1) and
klact(D2O)[D+]/s-1 were calculated from kobs[D+] = khyd(D2O)[D+] + klact(D2O)[D+] and Kequ/
6. Vercammen, K.; Glaus, M. A.; Van Loon, L. R. Radiochim. Acta 1999, 84, 221.
7. Vercammen, K.; Glaus, M. A.; Van Loon, L. R. Acta Chem. Scand. 1999, 53, 241.
8. Vercammen, K.; Glaus, M. A.; Van Loon, L. R. Radiochim. Acta 2001, 89, 393.
9. Warwick, P.; Evans, N.; Hall, T.; Vines, S. Radiochim. Acta 2003, 91, 233.
10. Warwick, P.; Evans, N.; Hall, T.; Vines, S. Radiochim. Acta 2004, 92, 897.
11. Warwick, P.; Evans, N.; Vines, S. Radiochim. Acta 2006, 94, 363.
12. Trinh, M. C.; Florent, J. C.; Monneret, C. J. Chem. Soc., Chem. Commun. 1987, 615.
13. Trinh, M. C.; Florent, J. C.; Monneret, C. Tetrahedron: Asymmetry 1988, 44, 6633.
14. Bertounesque, E.; Millal, F.; Meresse, P.; Monneret, C. Tetrahedron: Asymmetry
1998, 9, 2999.
D2O = klact(D2O)/khyd(D2O)
)
pD
kobs/s-1
Kequ/D2O
khyd(D2O)[D+]/s-1
klact(D2O)[D+]/s-1
1.49
1.67
2.37
6.73 E-5
5.17 E-5
3.48 E-6
5.13
3.22
2.21
1.09 E-5
1.23 E-5
1.41 E-6
5.63 E-5
3.94 E-5
2.40 E-6
15. Kim, J.; Hong, J. H. Carbohydr. Res. 2003, 338, 705.
16. Thomassigny, C.; Bennis, K.; Gelas, J. Synthesis 1997, 191, 1997.
17. Glaus, M. A.; Van Loon, L. R. Chemical Reactivity of Alpha-Isosaccharinic Acid in
Heterogeneous Alkaline Systems; Paul Scherrer Institut: Wurenlingen & Villigen;
2009.
18. Glaus, M. A.; van Loon, L. R.; Achatz, S.; Chodura, A.; Fischer, K. Anal. Chim. Acta
1999, 398, 111.
19. Greenfield, B. F., Harrison, W. N., Robertson, G. P., Somers, P. J., Spinder, W. W.,
Management, D. A. W., Eds.; AEA Technology: Harwell, 1993. p 1.
20. Greenfield, B. F.; Holtom, G. J.; Hurdus, M. H.; Okelly, N.; Pilkington, N. J.;
Rosevear, A.; Spindler, M. W.; Williams, S. J. Science 1995, 353, 1151.
21. Bontchev, R. P.; Moore, R.; Tucker, M. D.; Holt, K. Abstr. Pap. Am. Chem. Soc.
2004, 227, 031.
22. Bontchev, R. P.; Moore, R. C. Carbohydr. Res. 2004, 339, 801.
23. Bontchev, R. P.; Moore, R. C. Carbohydr. Res. 2004, 339, 2811.
24. Brown, P. L.; Allard, S.; Ekberg, C. J. Chem. Eng. Data. 2010, 55, 5207.
25. Cho, H.; Rai, D.; Hess, N. J.; Xia, Y. X.; Rao, L. F. J. Solution Chem. 2003, 32, 691.
26. Rai, D.; Hess, N. J.; Xia, Y. X.; Rao, L. F.; Cho, H. M.; Moore, R. C.; Van Loon, L. R. J.
Solution Chem. 2003, 32, 665.
to determine individual rate constants for the acid catalysed lac-
tone hydrolysis and lactone formation reaction and the value for
Kequ/D2O at that pH. The values calculated for khyd/D2O, klact/D2O
and Kequ/D2O at acidic pHs are presented in the Table 1.
Ekberg et al.48 have measured the same data set for the inter-
conversion of GISAL and GISAH at pH 1 (Kequ (6.60) khyd/H2O = 1.2
E-5 sꢀ1 and klact/H2O = 8.0 E-5 sꢀ1) however, caution should be prac-
tised in attempting a direct comparison of these figures because of
the different lactone structures and the requirement to consider
solvent isotope effects. Given the likely differences, it is surprising
that the numbers are similar. As stated above, the acid catalysed
hydrolysis reactions were very slow and when combined with
the limited availability of XISAH this meant that we were only able
to follow the reaction over a very limited pD range, that is, between
1 and 2.5. However, using the data presented in Table 1, we can get
an estimate for the second order rate constant for the acid cata-
27. Rai, D.; Rao, L. F.; Moore, D. A. Radiochim. Acta 1998, 83, 9.
28. Rai, D.; Yui, M.; Moore, D. A.; Rao, L. J. Solution Chem. 2009, 38, 1573.
29. Rai, D. P.; Rao, L. F.; Xia, Y. X. J. Solution Chem. 1998, 27, 1109.
30. Rao, L. F.; Garnov, A. Y.; Rai, D.; Xia, Y. X.; Moore, R. C. Radiochim. Acta 2004, 92,
575.
lysed hydrolysis reaction kD+ as 4.13 E-4 Mꢀ1 sꢀ1
.
31. Whistler, R.; Corbett, W. J. Am. Chem. Soc. 1956, 78, 1003.
32. Aspinall, G.; Carter, M.; Loss, M. J. Chem. Soc., Chem. Commun. 1956, 4807.
33. Kolmodin, H.; Samuelson, O. Sven. Papperstidn. 1971, 74, 301.
34. Kolmodin, H.; Samuelson, O. Sven. Papperstidn. 1973, 76, 71.
35. Niemela, K.; Alén, R.; Sjöström, E. Holzforschung 1985, 39, 167.
36. Niemelä, K.; Alén, R. In Analytical Methods in Wood Chemistry, Pulping, and
Papermaking; Sjöström, E., Alén, R., Eds.; Springer: Berlin, 1999. p 193.
37. Alén, R.; Valkonen, J. Acta Chem. Scand. 1995, 49, 536.
38. Alén, R. Acta Chem. Scand. Ser. B 1987, 41, 76.
39. Henderson, J. Am. J. Physiol. 1908, 21, 173.
40. Hasselbalch, K. Biochem. Z. 1917, 78, 112.
41. Shaw, P. B.; Robinson, G. F.; Rice, C. R.; Humphreys, P. N.; Laws, A. P. Carbohydr.
Res. 2012, 349, 6.
42. Glasoe, P.; Long, F. J. Am. Chem. Soc. 1960, 64, 188.
4. Conclusions
The pKa of XISAH has been measured using NMR methods and
the value (3.0 0.05) is lower than that measured for both alpha
and beta-GISAH using the same procedures. The rate constants
for hydrolysis of the lactone XISAL have been measured in both
acidic and basic solutions with the half-lives for reaction in acidic
and neutral solution being large. These slow transformation rates
will need to be considered when calculating the relative ratios of
the two species XISAH/L present in solution and this will be an
important consideration in calculating solubility and complexation
data for XISAH: between pH 4 and 7 an aqueous solution will take
several months to achieve complete equilibration.
43. Käkölä, J.; Alén, R.; Pakkanen, H.; Matilainen, R.; Lahti, K. J. Chromatogr. A 2007,
1139, 263.
44. Olson, A.; Hyde, J. J. Am. Chem. Soc. 1941, 63, 2459.
45. Grossert, J. S.; Cook, M. C.; White, R. L. Rapid Commun. Mass Spectrom. 2006, 20,
1511.
46. Stringer, M. B.; Bowie, J. H.; Eichinger, P. C. H.; Currie, G. J. J. Chem. Soc., Perkin
Trans. 2 1987, 385.
References
47. Blackburn, G.; Dodds, H. L. H. J. Chem. Soc., Perkin Trans. 2 1974, 377.
48. Ekberg, S.; Ekberg, C.; Albinsson, Y. J. Solution Chem. 2004, 33, 465.
1. Gakhokidze, R. Russ. Chem. Rev. 1980, 49, 222.
2. Sowden, J. C. In Advances in Carbohydrate Chemistry; Melville, L. W., Tipson, R. S.,
Eds.; Academic Press: New York; 1957; Vol. 12, p 35.