Sulfuric acid-catalyzed dehydration of sorbitol: Mechanistic study on preferential formation of 1,4-sorbitan

Article abstract of DOI:10.1246/bcsj.20150080

1,4-Sorbitan is a precursor to environmentally benign surfactants, which can be produced from biomass via sorbitol. Currently, sulfuric acid-catalyzed dehydration of sorbitol is the most widely used route for industrial synthesis of 1,4-sorbitan. In this work, we have studied the mechanism of the sorbitol dehydration by sulfuric acid. Our results show that both thermodynamic and kinetic parameters play significant roles to determine the yield of 1,4-sorbitan. Sorbitol preferentially forms an adduct with sulfuric acid, thereby inhibiting the subsequent dehydration of 1,4-sorbitan to isosorbide. Furthermore, a reaction mechanism is proposed for the dehydration reaction, which involves an S_N2 reaction on primary C1 of sorbitol attacked by OH of secondary C4.

Full text of DOI:10.1246/bcsj.20150080

Advance Publication Cover Page
Sulfuric Acid-Catalyzed Dehydration of Sorbitol: Mechanistic Study on
Preferential Formation of 1,4-Sorbitan
Mizuho Yabushita, Hirokazu Kobayashi, Abhijit Shrotri, Kenji Hara, Shogo Ito,
and Atsushi Fukuoka*
Advance Publication on the web April 16, 2015 by J-STAGE
doi:10.1246/bcsj.20150080
© 2015 The Chemical Society of Japan

Title: Sulfuric Acid-Catalyzed Dehydration of Sorbitol: Mechanistic Study on Preferential
Formation of 1,4-Sorbitan
Mizuho Yabushita,1,2 Hirokazu Kobayashi,1 Abhijit Shrotri,1 Kenji Hara,1 Shogo Ito,1,2 and
Atsushi Fukuoka*,1 1Catalysis Research Center, Hokkaido University, Kita 21 Nishi 10, Kita-ku,
Sapporo, Hokkaido 001-0021, Japan. 2Division of Chemical Sciences and Engineering,
Graduate School of Chemical Sciences and Engineering, Hokkaido University, Kita 13 Nishi 8,
Kita-ku, Sapporo, Hokkaido 060-8628, Japan.
Received: March 8, 2015;
E-mail: fukuoka@cat.hokudai.ac.jp
Synopsis
1,4-Sorbitan is a precursor to environmentally benign surfactants, which can be produced from
biomass via sorbitol. Currently, sulfuric acid-catalyzed dehydration of sorbitol is the most widely
used for industrial synthesis of 1,4-sorbitan. In this work, we have studied the mechanism of the
sorbitol dehydration by sulfuric acid. Our results show that both thermodynamic and kinetic
parameters play significant roles to determine the yield of 1,4-sorbitan. Sorbitol preferentially forms
an adduct with sulfuric acid, thereby inhibiting the subsequent dehydration of 1,4-sorbitan to
isosorbide. Furthermore, a reaction mechanism is proposed for the dehydration reaction, which
involves an SN2 reaction on primary C1 of sorbitol attacked by OH of secondary C4.
Introduction
Catalytic transformation of waste biomass to fuels and chemicals is a promising technology for
decreasing the emission of greenhouse gases causing anthropogenic climate change.1 Cellulose is the
most abundant naturally occurring biopolymer, which places the conversion of cellulose to valuable
chemicals as a primary step in bio-refineries. Many research groups have reported the hydrolytic
hydrogenation of cellulose to sorbitol over supported metal catalysts under H2 pressure.2 The
obtained sorbitol can be converted to useful chemicals such as 1,4-sorbitan (1,4-anhydrosorbitol) and
isosorbide (1,4:3,6-dianhydrosorbitol) (Scheme 1). In particular, 1,4-sorbitan is a raw material for
1

various environmentally benign surfactants3 [e.g., fatty acid esters (Span) and polysorbates (Tween)].
These compounds are extremely useful as emulsifiers, insecticides, and soft templates for material
synthesis.4 The emulsifiers can be used in food, cosmetic, and pharmaceutical industries owing to
their low-toxicity to human being.5 The current annual demand of sorbitan-derived surfactants is
more than 10,000 tons worldwide.4
Scheme 1. Dehydration of sorbitol.
At present, 1,4-sorbitan is produced commercially by sulfuric acid-catalyzed dehydration of
sorbitol with 58% yield.6 Use of heterogeneous catalysts for 1,4-sorbitan production can reduce
manufacture cost by eliminating problems associated with homogenous sulfuric acid catalyst.
Although 1,4-sorbitan can be produced at laboratory scale using solid acid catalysts or hot
compressed water, the inherent tendency of 1,4-sorbiton to undergo further dehydration to form
isosorbide reduces its yield.7 On the contrary, sulfuric acid can give 1,4-sorbitan by preferential
dehydration of sorbitol, but the mechanism is not well understood. The purpose of this work is to
elucidate the reason for a good yield of 1,4-sorbitan by kinetic study on the dehydration.
Results and Discussion
Here, this section is divided into three parts: (1) catalytic reaction and identification of products, (2)
kinetic study, and (3) reaction mechanism.
Catalytic reaction and identification of products. We conducted the dehydration of sorbitol at 403
K without using any solvent. Sorbitol (m.p. 368 K), 1,4-sorbitan (m.p. 385 K), and isosorbide (m.p.
336 K) are liquids at 403 K, which enabled us to perform the reaction under solvent-free conditions.
Use of sulfuric acid as catalyst resulted in 93% conversion of sorbitol for 45 min. The main product
was 1,4-sorbitan with 58% yield and 63% selectivity based on the conversion of sorbitol (93%)
(Scheme 2), which was in good agreement with the results reported earlier.6 The primary by-product
was isosorbide (16% yield) formed from successive dehydration of 1,4-sorbitan. 2,5-Mannitan
2

(2,5-anhydromannitol) was also formed as a by-product (3.4%). 2,5-Iditan (2,5-anhydroiditol) and a
mono-anhydrohexitol named AH1 were overlapped in the chromatograms (see Figures S1A and
S1C), and their total yield was 8.6%. The yield of another mono-anhydrohexitol labeled as AH2 was
2.0%. AH1 and AH2 were not fully identified, and characterization of the two compounds is
described in the next paragraph. A control experiment in the absence of catalyst provided no
conversion of sorbitol even after a long reaction time of 180 min, indicating that acid catalyst is
necessary for the dehydration of sorbitol.
Scheme 2. Dehydration of sorbitol by sulfuric acid. Reaction conditions: sorbitol 182 mg (1 mmol),
sulfuric acid 0.64 mg (6.5 μmol), 403 K, 45 min.
Identification of products is often difficult in a complex mixture of polyalcohols due to the
presence of various isomers. In this study, we used techniques such as 1D and 2D nuclear magnetic
resonance (NMR) along with liquid-chromatograph/mass spectroscopy (LC/MS) and
high-performance liquid chromatograph (HPLC). First, we isolated the products by HPLC equipped
with a fraction collector. Then, the isolated products were analyzed by comparing with commercial
standards in NMR, LC/MS, and HPLC. The 1H NMR spectrum, mass spectra, and HPLC charts of
the isolated 1,4-sorbitan were consistent with those of the standard sample (Figures 1 and S1).
Similarly, we identified other products such as isosorbide, 2,5-mannitan, and 2,5-iditan (see
Experimental and Figures S1–S4). Additionally, the LC/MS and NMR analyses revealed that the
mono-anhydrohexitol AH1 was formed (Figures S3 and S4), and 1H and 13C NMR indicated that this
product was not either 3,6-sorbitan (3,6-anhydrosorbitol) or 1,4-galactan (1,4-anhydrogalactitol).8
We also detected the other mono-anhydrohexitol AH2 by HPLC (Figure S1). Neither AH1 nor AH2
3

was identified as 1,5-sorbitan (1,5-anhydrosorbitol) or 1,5-mannitan (1,5-anhydromannitol) since the
retention times of AH1 and AH2 in HPLC were different from those of 1,5-sorbitan and 1,5-mannitan
(Figure S1). Other small peaks were also observed in HPLC, but they were not further investigated
as their yields were less than 1%. The yield of all identified products was determined using HPLC, in
which standard samples were used to calculate calibration factors. For 2,5-iditan, AH1, and AH2, the
calibration factor of 1,4-sorbitan was used to calculate their yields since regio- or stereoisomers
generally give similar calibration factors in a refractive index (RI) measurement.
Figure 1. Identification of 1,4-sorbitan produced from sorbitol. LC/MS spectra recorded by (A)
positive ion mode and (B) negative ion mode. The column used is Phenomenex Rezex
4

RPM-Monosaccharide Pb++. (A): 111 ([M + H − 3H2O]+), 129 ([M + H − 2H2O]+), 147 ([M + H −
H2O]+), 165 ([M + H]+). (B): 163 ([M − H]−), 196 ([M + O2]−). (C) 1H NMR spectra in D2O. The
assignment of 1H NMR peaks is summarized in Experimental. All spectra of 1,4-sorbitan obtained
from sorbitol are consistent with those of the standard sample.
Kinetic study of sorbitol dehydration by sulfuric acid. We studied the kinetics of sulfuric
acid-catalyzed dehydration of sorbitol. Figure 2 shows the time course of sorbitol dehydration by
sulfuric acid at 403 K. The amount of sorbitol decreased linearly to 7.4% in the first 45 min, which
was followed by a reduction in the rate of reaction. The apparent order of the reaction changed from
zeroth to first order at low concentrations of the substrate. This suggests that the reaction is
controlled by pre-equilibrium, which is the association equilibrium between the reactant and sulfuric
acid in this case. In Figure 2, the yield of 1,4-sorbitan was maximized (58%) at 45 min. Contrastingly,
only a small amount of isosorbide was formed in the first 30 min regardless of the accumulation of
1,4-sorbitan. Thereafter, the yield of isosorbide increased at an accelerated pace to reach 72% in 180
min. Since the conversion rate of 1,4-sorbitan to isosorbide increased after a large portion of sorbitol
was consumed, we can postulate that sorbitol forms an adduct with sulfuric acid more preferentially
than 1,4-sorbitan.
Figure 2. Time course of sorbitol dehydration by sulfuric acid at 403 K. Reaction conditions:
sorbitol 182 mg (1 mmol), sulfuric acid 0.64 mg (6.5 μmol). Data points show experimental results
and lines are simulation curves based on eqs 4–9. The simulation curves were drawn by considering
5

preferential adsorption of sorbitol over 1,4-sorbitan and isosorbide (K1/K2 = 4.3 and K1/K3 = 4.5).
Legend: black diamonds (♦) = amount of unreacted sorbitol; red circles (●) = yield of 1,4-sorbitan;
and blue triangles (▲) = yield of isosorbide. The standard deviations for the difference between
experimental data and simulation results were calculated to be 3.9% for sorbitol, 2.2% for
1,4-sorbitan, and 1.1% for isosorbide.
We simulated the time course of sorbitol dehydration by assuming the Michaelis-Menten type
mechanism involving a pre-equilibrium of association and a subsequent dehydration step (Scheme 3),
where k1 and k2 are rate constants for dehydration of sorbitol and 1,4-sorbitan, respectively. Whereas,
k3, k4, and k5 are rate constants for side reactions of sorbitol, 1,4-sorbitan, and isosorbide. The rates of
side reactions were estimated from the carbon balance of sorbitol, 1,4-sorbitan, and isosorbide. The
equilibrium constants for association of sulfuric acid with sorbitol, 1,4-sorbitan, and isosorbide are
denoted as K1, K2, and K3, respectively. The equilibrium constants are presented below (eqs 1‒3).
Rate constant /10-2 min-1
K1/K2
K1/K3
k1
k2
k3
k4
k5
4.7
2.7
1.1
0.38
<0.01
4.3
4.5
Scheme 3. Reaction schemes of sulfuric acid-catalyzed dehydration of sorbitol. Rate constants and
ratios of association equilibrium constants at 403 K were determined by the fitting simulations based
on eqs 4–9 (Figure 2).
[
Sorbitol⋅H2SO4
]
K1 =
K2 =
(1)
[
Sorbitol
]
(
[
H2SO4
]
−
[
Sorbitol⋅H2SO4 Sorbitan ⋅H2SO4
]
−
[
]
−
[
Isosorbide⋅H2SO4
]
)
0
[
Sorbitan ⋅H2SO4
]
(2)
[
Sorbitan
]
(
[
H2SO4
]
−
[
Sorbitol⋅H2SO4 Sorbitan ⋅H2SO4
]
−
[
]
−
[
Isosorbide⋅H2SO4
]
)
0
6

[
Isosorbide⋅H2SO4
]
K3 =
(3)
[
Isosorbide
]
(
[
H2SO4
]
−
[
Sorbitol⋅H2SO4 Sorbitan ⋅H2SO4
]
−
[
]
−
[
Isosorbide⋅H2SO4
]
)
0
where [H2SO4]0 is the initial concentration of sulfuric acid. [Sorbitol], [Sorbitan], and [Isosorbide]
are the concentrations of sorbitol, 1,4-sorbitan, and isosorbide, respectively. [Sorbitol · H2SO4],
[Sorbitan · H2SO4], and [Isosorbide · H2SO4] are the concentrations of sulfuric acid adducts with
sorbitol, 1,4-sorbitan, and isosorbide. Accordingly,
K
1[Sorbitol][H2SO4 ]
0
[
Sorbitol⋅H2SO4
]
=
(4)
(5)
1+ K
[
1 Sorbitol
]
+ K2
[
Sorbitan
]
+ K3
[
Isosorbide
]
K2
Sorbitol
[
Sorbitan][H2SO4 ]
0
[
Sorbitan ⋅H2SO4 =
]
1+ K1
[
]
+ K2
[
Sorbitan
]
+ K3
[
Isosorbide
]
K3
[
Isosorbide][H2SO4
]
0
[
Isosorbide⋅H2SO4
]
=
(6)
1+ K
[
1 Sorbitol
]
+ K2
[
Sorbitan
]
+ K3
[
Isosorbide
]
The reaction rates corresponding to each reactant can be represented by using reaction time t as
d
d
d
[
Sorbitol
]
= −
(7)
(
k1 + k3
)
[
Sorbitol⋅H2SO4
]
dt
[
Sorbitan
]
= k
(8)
[
1 Sorbitol⋅H2SO4
]
−
(
k2 + k4
)
[
Sorbitan ⋅H2SO4
]
dt
[
Isosorbide
]
= k2
(9)
[
Sorbitan ⋅H2SO4
]
− k5
[
Isosorbide⋅H2SO4
]
dt
The simulation curves (lines) based on eqs 4‒9 fit well with the experimental data (dots) in Figure 2.
According to the calculation, the rate constant k1 (4.7 × 10-2 min-1) was only 1.7 times larger than k2
(2.7 × 10-2 min-1) (Scheme 3). Furthermore, k3 (1.1 × 10-2 min-1) and k4 (3.8 × 10-3 min-1) were lower
than k1, which was consistent with the low yields of by-products such as 2,5-mannitan and 2,5-iditan.
The rate constant k5 was <1.0 × 10-4 min-1, indicating high stability of isosorbide under the reaction
conditions.
7

For the association equilibrium constants, the K1/K2 and K1/K3 ratios were estimated to be 4.3
and 4.5, respectively. This indicates that the association of sulfuric acid with sorbitol is 4.3 and 4.5
times stronger than that with 1,4-sorbitan and isosorbide. If this preferential association did not occur
(e.g., K1 = K2 = K3), the maximum yield of 1,4-sorbitan would be at most 40% (Figure S5). In this
case, in order to achieve the original yield of 1,4-sorbitan (58%), the rate of the dehydration of
sorbitol to 1,4-sorbitan needs to be 3.2 times larger than that of the subsequent transformation of
1,4-sorbitan. In other words, the value of k1/(k2 + k4) should be greater than 3.2. In reality, the
calculated value for k1/(k2 + k4) is only 1.2 as both sorbitol and 1,4-sorbitan dehydration are similar
acid-catalyzed reactions. Therefore, we conclude that the inhibition of association between sulfuric
acid and 1,4-sorbitan due to the large K1/K2 ratio prevents the successive dehydration of 1,4-sorbitan.
The above results support the hypothesis by Takagaki et al. that molecular recognition is the key for
selective formation of 1,4-sorbitan.7j
Reaction mechanism for dehydration of sorbitol by sulfuric acid. As mentioned in the previous
section, sorbitol undergoes dehydration after the formation of adduct with sulfuric acid. Here, we
discuss the mechanism of the dehydration based on the product distribution (Scheme 2). Five types
of mechanisms are possible for the 1,4-dehydration of sorbitol (Scheme 4): (1) SN2 reaction at
primary C1 attacked by a secondary OH of C4,9 (2) SN1 reaction at C1 attacked by the OH of C4, (3)
SN2 reaction at secondary C4 attacked by a primary OH of C1, (4) SN1 reaction at C4 attacked by the
OH of C1, and (5) tandem SN2-SN2 reactions caused by a vicinal OH and a terminal OH.10 Taking
into account the higher selectivity of 1,4-sorbitan, the mechanism 1 is the most probable pathway for
the following reasons. In the mechanism 1, the SN2 reaction at C1 is favorable with less steric
hindrance and retention of the configuration of sorbitol. This 1,4-dehydration selectively provides
1,4-sorbitan. In the mechanism 2, the formation of primary carbenium ion at C1 is energetically
unfavorable. According to the density functional theory (DFT) calculations for dehydration of sugar
alcohol by Nimlos et al., 1,2-hydride shift simultaneously occurs to form secondary carbenium ion,
which is more stable than primary carbenium ion.11 Therefore, it is less likely that the reaction
proceeds through mechanism 2. The SN2 reaction at C4 (mechanism 3) should result in the formation
of 1,4-galactan as a major product. As this product has not been detected, this mechanism can be
excluded. Similarly, the mechanism 4 is implausible as the formation of secondary carbenium ions
should yield a mixture of 1,4-sorbitan and 1,4-galactan, but the latter was not observed in our
8

reaction. In the mechanism 5, if the first SN2 reaction occurs at C3 or C5, the stereochemistry of
sorbitol is maintained to produce 1,4-sorbitan as shown in Scheme 4. However, the first SN2 reaction
is possible by OH group on C4 to C3 (mechanism 5') or C5 (mechanism 5''). Such reactions would
result in the formation of 3,6-sorbitan (L-isomer) or 3,6-talitan (3,6-anhydrotalitol, L-isomer), but
these were not detected in our products. Hence, we propose that the selective 1,4-dehydration of
sorbitol by sulfuric acid occurs by the mechanism 1, viz., the SN2 reaction at C1.
Scheme 4. Possible reaction pathways for 1,4-dehydration of sorbitol.
9

Conclusions
Sulfuric acid-catalyzed dehydration of sorbitol produced 1,4-sorbitan in 58% yield with 63%
selectivity at 403 K. Kinetic study shows that both sorbitol and 1,4-sorbitan undergo dehydration by
acid catalysts as indicated by similar rate constants (k1/k2 = 1.7). However, the ratio of association
equilibrium constants (K1/K2 = 4.3) is high. Therefore, the association of sorbitol with sulfuric acid is
substantially more favorable than that of 1,4-sorbitan with sulfuric acid, and further dehydration of
1,4-sorbitan to isosorbide is effectively inhibited. Accordingly, we believe that control of the
association equilibrium constant (K) is essential in addition to an increase in the rate constant (k) to
maximize the yield of 1,4-sorbitan. This insight is useful for designing novel catalysts for the
selective synthesis of anhydropolyols.
Experimental
Reagents. Sulfuric acid was obtained from Wako Pure Chemical Industries, sorbitol and isosorbide
from Tokyo Chemical Industry, 1,4-sorbitan and 1,5-sorbitan from Toronto Research Chemicals,
1,5-mannitan and 2,5-mannitan from Sigma-Aldrich. Distilled water and deuterated solvents
[deuterium oxide and dimethyl sulfoxide-d6 (DMSO-d6)] were supplied by Wako Pure Chemical
Industries and Acros Organics, respectively.
Dehydration of sorbitol. Sulfuric acid (13 mg) was dissolved in diethyl ether (10 mL), and then
powder of sorbitol (3.64 g) was added to the solution. After drying diethyl ether under vacuum,
sulfuric acid-impregnated sorbitol was obtained. The dehydration reaction was conducted without
using solvent in an eggplant flask (50 mL). The melting points of sorbitol, 1,4-sorbitan, and
isosorbide are 368, 385, and 336 K, respectively, which have enabled us to conduct the dehydration
reaction in the absence of solvents at 403 K. The sulfuric acid-impregnated sorbitol [containing
sorbitol 182 mg (1 mmol) and sulfuric acid 0.64 mg (6.5 μmol)] was charged into the flask. The flask
was immersed in an oil bath at 403 K. The volume of reactant was 1.32 × 10-4 L [calculated from the
amount of sorbitol charged (182 mg) and the density of sorbitol at 403 K (0.726 ×10-6 L mg-1)12]
during the reaction at 403 K, and this value was used for the simulation based on eqs 4–9. After the
reaction, 20 mL of water was added to the mixture at room temperature to dissolve the products. The
water-soluble compounds were analyzed by HPLC [Shimadzu LC10-ATVP, RI and ultraviolet (210
nm) detectors] with a Shodex SUGAR SH1011 column (ø8 × 300 mm, mobile phase: water 0.5 mL
10

min-1, 323 K) and a Phenomenex Rezex RPM-Monosaccharide Pb++ column (ø7.8 × 300 mm,
mobile phase: water 0.6 mL min-1, 343 K). Yields of products and sorbitol recovery were determined
by an absolute calibration method.
Product identification. The products were identified by means of NMR methods [JEOL,
JNM-ECX-400 (1H 400 MHz, 13C 100 MHz) and JNM-ECX-600 (1H 600 MHz, 13C 150 MHz)
including 1H NMR spectroscopy, proton-decoupled 13C NMR spectroscopy, distortionless
enhancement by polarization transfer (DEPT), and 13C−1H heteronuclear multiple quantum
coherence (HMQC) technique, Figures 1, S4, and S5]. LC/MS [Thermo Fisher Scientific, LCQ Fleet,
atmospheric pressure chemical ionization (APCI), Figures 1, S2, and S3] was conducted under the
same conditions as described above for HPLC using the Phenomenex Rezex RPM-Monosaccharide
Pb++ column. The structures of identified products (e.g., 1,4-sorbitan, isosorbide, 2,5-mannitan, and
2,5-iditan) are shown in Scheme 2. Mono-anhydrohexitols, denoted as AH1 and AH2, were also
observed by HPLC (Figure S1). Although 2,5-iditan and AH1 were not separated by HPLC, they
were detected by NMR (Figure S4).
1,4-Sorbitan: 1H NMR (400 MHz, D2O): δ 4.27 (1H, d, J = 4.0 Hz, H–C2), 4.21 (1H, d, J = 2.8
Hz, H–C3), 4.13 (1H, dd, J = 10, 4.0 Hz, H–C1), 3.90 (1H, dd, J = 8.8, 2.8 Hz, H–C4), 3.85 (1H,
ddd, J = 8.8, 6.0, 2.8 Hz, H–C5), 3.78 (1H, dd, J = 12, 2.8 Hz, H–C6), 3.70 (1H, d, J = 10 Hz, H–C1),
3.61 (1H, dd, J = 12, 6.0 Hz, H–C6). H–C1 and H–C6 were distinguished by means of 1H NMR in
DMSO-d6 solvent. In DMSO-d6, the peaks for H–C6 are observed as ddd due to coupling with one
proton of the hydroxyl group in addition to two aliphatic protons, whereas the peaks of H–C1 are in a
dd form due to coupling with one proton of the hydroxyl group in addition to one aliphatic proton.
LC/MS: [M + H]+ calculated 165, found 165 (reaction product), found 165 (commercial standard);
[M − H]− calculated 163, found 163 (reaction product), found 163 (commercial standard).
Isosorbide: 1H NMR (400 MHz, D2O): δ 4.64 (1H, dd, J = 5.2, 4.4 Hz, H–C4), 4.49 (1H, d, J =
4.4 Hz, H–C3), 4.39 (1H, ddd, J = 7.0, 7.0, 5.2 Hz, H–C5), 4.32 (1H, d, J = 3.2 Hz, H–C2),
3.89‒3.96 (2H, H–C1 and H–C6), 3.87 (1H, dd, J = 10.4, 3.2 Hz, H–C1), 3.49 (1H, dd, J = 9.0, 7.2
Hz, H–C6). Some of the 1H NMR peaks at 3.89‒3.96 and 3.87 ppm were overlapped with the peaks
of AH1 (Figure S4A). The 13C NMR and DEPT peaks of isosorbide in Figure S4B could not be
distinguished from noise due to its low concentration.
2,5-Mannitan: LC/MS: [M + H]+ calculated 165, found 165 (reaction product), found 165
11

(commercial standard); [M − H]− calculated 163, found 163 (reaction product), found 163
(commercial standard).
2,5-Iditan: 1H NMR (600 MHz, D2O): δ 4.21–4.25 (4H, H–C2, H–C3, H–C4, and H–C5), 3.80
(2H, dd, J = 11.4, 4.8 Hz, H–C1 and H–C6), 3.73 (2H, dd, J = 11.4, 7.2 Hz, H–C1 and H–C6). 13C
NMR (150 MHz, D2O): δ 81.3 (CH, C3 and C4), 77.5 (CH, C2 and C5), 61.0 (CH2, C1 and C6). The
1H NMR peaks at 4.21–4.25, 3.80, and 3.73 ppm were overlapped with those of AH1 (Figure S4A).
The assignment of the 13C NMR peaks is based on reference 13. LC/MS for reaction product [M +
H]+ calculated 165, found 165; [M − H]− calculated 163, found 163.
Mono-anhydrohexitol (AH1): 1H NMR (600 MHz, D2O): δ 4.29 (1H, d, J = 4.2 Hz), 4.15 (1H,
dd, J = 10.5, 4.2 Hz), 3.92 (1H, dd, J = 9.0, 2.4 Hz), 3.88 (1H, ddd, J = 9.0, 6.0, 3.0 Hz), 3.64 (1H,
dd, J = 12, 6.0 Hz). 13C NMR (150 MHz, D2O): δ 80.7 (CH), 77.2 (CH), 76.7 (CH), 74.3 (CH2), 69.9
(CH), 64.6 (CH2). Several 1H NMR peaks were overlapped with those of 2,5-iditan and isosorbide
(Figure S4A). The chemical shifts of the 1H NMR peaks indicate that this unidentified compound is
not 3,6-sorbitan or 1,4-galactan.8 The retention times of AH1 in HPLC analyses are different from
those of 1,5-sorbitan or 1,5-mannitan (Figure S1). LC/MS for reaction product [M + H]+ found 165;
[M − H]− found 163.
Mono-anhydrohexitol (AH2): The retention times of AH2 in HPLC analyses are different from
those of 1,5-sorbitan or 1,5-mannitan (Figure S1). LC/MS for reaction product [M + H]+ found 165;
[M − H]− found 163.
The authors would like to thank Dr. Y. Koyama (Hokkaido University) for the discussion of
NMR assignments and Dr. Y. Yamakoshi (Hokkaido Research Organization) for the LC/MS analysis.
This work was supported by a Grant-in-Aid for the Japan Society for the Promotion of Science
(JSPS) Research Fellows.
Supporting Information: HPLC data for products in sorbitol dehydration by sulfuric acid, LC/MS
and NMR spectra of products, and simulation curves with the assumption of K1 = K2 = K3. This
material is available free of charge on the web at https://www.jstage.jst.go.jp/browse/bcsj.
12

References
1
a) A. Corma, S. Iborra, A. Velty, Chem. Rev. 2007, 107, 2411. b) M. E. Himmel, S.-Y. Ding, D.
K. Johnson, W. S. Adney, M. R. Nimlos, J. W. Brady, T. D. Foust, Science 2007, 315, 804. c) J. S.
Luterbacher, D. M. Alonso, J. A. Dumesic, Green Chem. 2014, 16, 4816. d) M. Yabushita, H.
Kobayashi, A. Fukuoka, Appl. Catal. B: Environ. 2014, 145, 1.
2
a) A. Fukuoka, P. L. Dhepe, Angew. Chem., Int. Ed. 2006, 45, 5161. b) C. Luo, S. Wang, H. Liu,
Angew. Chem., Int. Ed. 2007, 46, 7636. c) W. Deng, X. Tan, W. Fang, Q. Zhang, Y. Wang, Catal.
Lett. 2009, 133, 167. d) V. Jollet, F. Chambon, F. Rataboul, A. Cabiac, C. Pinel, E. Guillon, N.
Essayem, Green Chem. 2009, 11, 2052. e) J. Geboers, S. Van de Vyver, K. Carpentier, K. de
Blochouse, P. Jacobs, B. Sels, Chem. Commun. 2010, 46, 3577. f) I. A. Ignatyev, C. V. Doorslaer,
P. G. N. Mertens, K. Binnemans, D. E. De Vos, ChemSusChem 2010, 3, 91. g) R. Palkovits, K.
Tajvidi, A. M. Ruppert, J. Procelewska, Chem. Commun. 2011, 47, 576. h) H. Kobayashi, Y. Ito,
T. Komanoya, Y. Hosaka, P. L. Dhepe, K. Kasai, K. Hara, A. Fukuoka, Green Chem. 2011, 13,
326. i) H. Kobayashi, H. Matsuhashi, T. Komanoya, K. Hara, A. Fukuoka, Chem. Commun.
2011, 47, 2366. j) J. Geboers, S. Van de Vyver, K. Carpentier, P. Jacobs, B. Sels, Green Chem.
2011, 13, 2167. k) M. Liu, W. Deng, Q. Zhang, Y. Wang, Y. Wang, Chem. Commun. 2011, 47,
9717. l) Y. Ogasawara, S. Itagaki, K. Yamaguchi, N. Mizuno, ChemSusChem 2011, 4, 519. m) J.
Pang, A. Wang, M. Zheng, Y. Zhang, Y. Huang, X. Chen, T. Zhang, Green Chem. 2012, 14, 614.
n) S. Van de Vyver, J. Geboers, W. Schutyser, M. Dusselier, P. Eloy, E. Dornez, J. W. Seo, C. M.
Courtin, E. M. Gaigneaux, P. A. Jacobs, B. F. Sels, ChemSusChem 2012, 5, 1549. o) J. W. Han,
H. Lee, Catal. Commun. 2012, 19, 115. p) A. Shrotri, A. Tanksale, J. N. Beltramini, Catal. Sci.
Technol. 2012, 2, 1852. q) M. Käldström, N. Kumar, M. Tenho, M. V. Mokeev, Y. E.
Moskalenko, D. Y. Murzin, ACS Catal. 2012, 2, 1381. r) D. Reyes-Luyanda, J. Flores-Cruz, P. J.
Morales-Pérez, L. G. Encarnación-Gómez, F. Shi, P. M. Voyles, N. Cardona-Martínez, Top.
Catal. 2012, 55, 148. s) J. Hilgert, N. Meine, R. Rinaldi, F. Schüth, Energy Environ. Sci. 2013, 6,
92. t) T. Deng, H. Liu, Green Chem. 2013, 15, 116. u) T. Komanoya, H. Kobayashi, K. Hara,
W.-J. Chun, A. Fukuoka, ChemCatChem 2014, 6, 230. v) H. Kobayashi, Y. Hosaka, K. Hara, B.
Feng, Y. Hirosaki, A. Fukuoka, Green Chem. 2014, 16, 637. w) G. Liang, L. He, H. Cheng, W. Li,
X. Li, C. Zhang, Y. Yu, F. Zhao, J. Catal. 2014, 309, 468. x) D. Wang, W. Niu, M. Tan, M. Wu,
X. Zheng, Y. Li, N. Tsubaki, ChemSusChem 2014, 7, 1398.
13

3
4
5
6
X. Li, Y. Su, X. Zhou, X. Mo, Colloids Surf. B 2009, 69, 221.
H. Kobayashi, A. Fukuoka, Green Chem. 2013, 15, 1740.
T. Yoshioka, B. Sternberg, A. T. Florence, Int. J. Pharm. 1994, 105, 1.
a) S. Soltzberg, U.S. Patent 2,390,395, 1945. b) S. Soltzberg, R. M. Goepp Jr., W. Freudenberg,
J. Am. Chem. Soc. 1946, 68, 919. c) M. A. Andrews, K. K. Bhatia, P. J. Fagan, U.S. Patent
6,689,892, 2004.
7
a) C. Montassier, J. C. Menezo, J. Moukolo, J. Naja, L. C. Hoang, J. Barbier, J. P. Boitiaux, J.
Mol. Catal. 1991, 79, 65. b) A. J. Sanborn, U.S. Patent 7,420,067, 2008. c) K. M. Moore, A. J.
Sanborn, P. Bloom, U.S. Patent 7,439,352, 2008. d) J. E. Holladay, J. Hu, X. Zhang, Y. Wang,
U.S. Patent 7,772,412, 2010. e) J. Xia, D. Yu, Y. Hu, B. Zou, P. Sun, H. Li, H. Huang, Catal.
Commun. 2011, 12, 544. f) A. Liu, C. Luckett, U.S. Patent 7,982,059, 2011. g) A. Yamaguchi, N.
Hiyoshi, O. Sato, M. Shirai, Green Chem. 2011, 12, 873. h) M. Rose, R. Palkovits,
ChemSusChem 2012, 5, 167. i) N. A. Khan, D. K. Mishra, I. Ahmed, J. W. Yoon, J.-S. Hwang, S.
H. Jhung, Appl. Catal. A: Gen. 2013, 452, 34. j) Y. Morita, S. Furusato, A. Takagaki, S. Hayashi,
R. Kikuchi, S. T. Oyama, ChemSusChem 2014, 7, 748. k) H. Kobayashi, H. Yokoyama, B. Feng,
A. Fukuoka, Green Chem., doi:10.1039/C5GC00319A.
8
9
R. Wieneke, S. Klein, A. Geyer, E. Loos, Carbohydr. Res. 2007, 342, 2757.
L. Došen-Mićović, Ž. Čeković, J. Phys. Org. Chem. 1998, 11, 887.
10 R. Barker, J. Org. Chem. 1970, 35, 461.
11 M. R. Nimlos, S. J. Blanksby, X. Qian, M. E. Himmel, D. K. Johnson, J. Phys. Chem. A 2006,
110, 6145.
12 M. Naoki, K. Ujita, S. Kashima, J. Phys. Chem. 1993, 97, 12356.
13 K. Bock, C. Pedersen, H. Thøgersen, Acta Chem. Scand. B 1981, 35, 441.
14

Scheme captions
Scheme 1. Dehydration of sorbitol.
Scheme 2. Dehydration of sorbitol by sulfuric acid. Reaction conditions: sorbitol 182 mg (1 mmol),
sulfuric acid 0.64 mg (6.5 μmol), 403 K, 45 min.
Scheme 3. Reaction schemes of sulfuric acid-catalyzed dehydration of sorbitol. Rate constants and
ratios of association equilibrium constants at 403 K were determined by the fitting
simulations based on eqs 4–9 (Figure 2).
Scheme 4. Possible reaction pathways for 1,4-dehydration of sorbitol.
Figure captions
Figure 1. Identification of 1,4-sorbitan produced from sorbitol. LC/MS spectra recorded by (A)
positive ion mode and (B) negative ion mode. The column used is Phenomenex Rezex
RPM-Monosaccharide Pb++. (A): 111 ([M + H − 3H2O]+), 129 ([M + H − 2H2O]+), 147
([M + H − H2O]+), 165 ([M + H]+). (B): 163 ([M − H]−), 196 ([M + O2]−). (C) 1H NMR
spectra in D2O. The assignment of 1H NMR peaks is summarized in Experimental. All
spectra of 1,4-sorbitan obtained from sorbitol are consistent with those of the standard
sample.
Figure 2. Time course of sorbitol dehydration by sulfuric acid at 403 K. Reaction conditions:
sorbitol 182 mg (1 mmol), sulfuric acid 0.64 mg (6.5 μmol). Data points show
experimental results and lines are simulation curves based on eqs 4–9. The simulation
curves were drawn by considering preferential adsorption of sorbitol over 1,4-sorbitan
and isosorbide (K1/K2 = 4.3 and K1/K3 = 4.5). Legend: black diamonds (♦) = amount of
unreacted sorbitol; red circles (●) = yield of 1,4-sorbitan; and blue triangles (▲) = yield
of isosorbide. The standard deviations for the difference between experimental data and
simulation results were calculated to be 3.9% for sorbitol, 2.2% for 1,4-sorbitan, and
1.1% for isosorbide.
15

Scheme 1. Dehydration of sorbitol.
(M. Yabushita, H. Kobayashi, A. Shrotri, K. Hara, S. Ito, A. Fukuoka)
16

Scheme 2. Dehydration of sorbitol by sulfuric acid. Reaction conditions: sorbitol 182 mg (1 mmol),
sulfuric acid 0.64 mg (6.5 μmol), 403 K, 45 min.
(M. Yabushita, H. Kobayashi, A. Shrotri, K. Hara, S. Ito, A. Fukuoka)
17

Rate constant /10-2 min-1
K1/K2
K1/K3
k1
k2
k3
k4
k5
4.7
2.7
1.1
0.38
<0.01
4.3
4.5
Scheme 3. Reaction schemes of sulfuric acid-catalyzed dehydration of sorbitol. Rate constants and
ratios of association equilibrium constants at 403 K were determined by the fitting simulations based
on eqs 4–9 (Figure 2).
(M. Yabushita, H. Kobayashi, A. Shrotri, K. Hara, S. Ito, A. Fukuoka)
18

Scheme 4. Possible reaction pathways for 1,4-dehydration of sorbitol.
(M. Yabushita, H. Kobayashi, A. Shrotri, K. Hara, S. Ito, A. Fukuoka)
19

Figure 1. Identification of 1,4-sorbitan produced from sorbitol. LC/MS spectra recorded by (A)
positive ion mode and (B) negative ion mode. The column used is Phenomenex Rezex
RPM-Monosaccharide Pb++. (A): 111 ([M + H − 3H2O]+), 129 ([M + H − 2H2O]+), 147 ([M + H −
H2O]+), 165 ([M + H]+). (B): 163 ([M − H]−), 196 ([M + O2]−). (C) 1H NMR spectra in D2O. The
assignment of 1H NMR peaks is summarized in Experimental. All spectra of 1,4-sorbitan obtained
from sorbitol are consistent with those of the standard sample.
(M. Yabushita, H. Kobayashi, A. Shrotri, K. Hara, S. Ito, A. Fukuoka)
20

Figure 2. Time course of sorbitol dehydration by sulfuric acid at 403 K. Reaction conditions:
sorbitol 182 mg (1 mmol), sulfuric acid 0.64 mg (6.5 μmol). Data points show experimental results
and lines are simulation curves based on eqs 4–9. The simulation curves were drawn by considering
preferential adsorption of sorbitol over 1,4-sorbitan and isosorbide (K1/K2 = 4.3 and K1/K3 = 4.5).
Legend: black diamonds (♦) = amount of unreacted sorbitol; red circles (●) = yield of 1,4-sorbitan;
and blue triangles (▲) = yield of isosorbide. The standard deviations for the difference between
experimental data and simulation results were calculated to be 3.9% for sorbitol, 2.2% for
1,4-sorbitan, and 1.1% for isosorbide.
(M. Yabushita, H. Kobayashi, A. Shrotri, K. Hara, S. Ito, A. Fukuoka)
21

Graphical Abstract
Sulfuric Acid-Catalyzed Dehydration of Sorbitol: Mechanistic Study on Preferential Formation
of 1,4-Sorbitan
Mizuho Yabushita, Hirokazu Kobayashi, Abhijit Shrotri, Kenji Hara, Shogo Ito,
and Atsushi Fukuoka
Sulfuric acid produces 1,4-sorbitan from sorbitol by controlling thermodynamics and kinetics.
Sorbitol preferentially forms an adduct with sulfuric acid, which suppresses further dehydration of
1,4-sorbitan to isosorbide.
27