Biocatalytic synthesis of ultra-long-chain fatty acid sugar alcohol monoesters

Article abstract of DOI:10.1039/c5gc00695c

An array of ultra-long-chain fatty acid sugar alcohol monoesters, with behenic acid as an acyl representative and sugar alcohols altered from ethylene glycol to glycerol, erythritol, pentaerythritol, arabitol, xylitol and sorbitol, were enzymatically synthesized in high purity and selectivity. The molecular structures of the synthetic compounds were confirmed by ¹H NMR, FT-IR and MS analysis, and the thermal properties were primarily characterized by DSC analysis. The molecular packing and thermal properties of synthetic sugar alcohol monobehenates (SAMBs) were investigated by Temp-Ramp-FT-IR. For in vivo application purposes, the enzymatic lipolysis of synthesized SAMBs was examined by a PPL (porcine pancreatic lipase)-mediated in vitro digestion test, and improved resistance of most SAMBs to enzymatic lipolysis, in comparison to glycerol monopalmitate, was observed. FT-IR spectroscopic analysis of the thermotropic phase transitions of the synthetic SAMBs indicated that the thermal collapse temperatures do not vary significantly as the polar head alters, suggesting their thermostabilities are largely governed by hydrophobic interactions among the alkanyl chains, while the size, properties and volume of the polar heads may determine the packing patterns. Systemic mapping of the structure-property-function relationship of SAMBs revealed the potential of these compounds for multipurpose applications. Ethylene glycol and glycerol monobehenates enable orthorhombic packing, and could find applications in cosmetic formulation, whereas sorbitol monobehenate is capable of forming stable surfactant-free nanoparticles, which could be excellent excipients for solid lipid nanoparticles for use as delivery cargo for drugs and food ingredients.

Full text of DOI:10.1039/c5gc00695c

Green Chemistry
View Article Online
View Journal
PAPER
Biocatalytic synthesis of ultra-long-chain fatty acid
sugar alcohol monoesters
Cite this: DOI: 10.1039/c5gc00695c
Wei Wei,a,b Fengqin Feng,a Bianca Perez,b Mingdong Dongc and Zheng Guo*b
An array of ultra-long-chain fatty acid sugar alcohol monoesters, with behenic acid as an acyl representa-
tive and sugar alcohols altered from ethylene glycol to glycerol, erythritol, pentaerythritol, arabitol, xylitol
and sorbitol, were enzymatically synthesized in high purity and selectivity. The molecular structures of the
synthetic compounds were confirmed by 1H NMR, FT-IR and MS analysis, and the thermal properties
were primarily characterized by DSC analysis. The molecular packing and thermal properties of synthetic
sugar alcohol monobehenates (SAMBs) were investigated by Temp-Ramp-FT-IR. For in vivo application
purposes, the enzymatic lipolysis of synthesized SAMBs was examined by a PPL (porcine pancreatic
lipase)-mediated in vitro digestion test, and improved resistance of most SAMBs to enzymatic lipolysis, in
comparison to glycerol monopalmitate, was observed. FT-IR spectroscopic analysis of the thermotropic
phase transitions of the synthetic SAMBs indicated that the thermal collapse temperatures do not vary sig-
nificantly as the polar head alters, suggesting their thermostabilities are largely governed by hydrophobic
interactions among the alkanyl chains, while the size, properties and volume of the polar heads may
determine the packing patterns. Systemic mapping of the structure–property–function relationship of
SAMBs revealed the potential of these compounds for multipurpose applications. Ethylene glycol and gly-
cerol monobehenates enable orthorhombic packing, and could find applications in cosmetic formulation,
whereas sorbitol monobehenate is capable of forming stable surfactant-free nanoparticles, which
could be excellent excipients for solid lipid nanoparticles for use as delivery cargo for drugs and
food ingredients.
Received 31st March 2015,
Accepted 20th April 2015
DOI: 10.1039/c5gc00695c
www.rsc.org/greenchem
competition, is the technical feasibility, efficacy and atom-
economy of the synthesis pathways and production routes
Introduction
leading to the target products.8–10 In this respect, as in vitro
biosynthesis workshops, enzyme-mediated reaction systems
are an attractive tool because of their mild reaction conditions,
simple post-processing, high selectivity, smaller amount of
waste, and production flexibility.11–13
Green chemistry may contribute to the minimization of the
environmental impact of production activities from both the
origins and the processing and preparation of industrially rele-
vant products.1,2 Due to depleting petroleum resources, build-
ing blocks derived from annually renewable feedstocks for the
synthesis and production of bioproducts relevant to commer-
cial applications are becoming increasingly important.3–5
However, aside from “greenness”, functionality and product
performance will be important factors for most bio-based pro-
ducts under development because of the considerations of
cost-effectiveness and market pressure.3,6,7 Another important
factor, which is linked to environmental impact and market
Sugar alcohol fatty acid esters are a group of naturally
occurring compounds found ubiquitously in microorganisms
and plants,14–16 which consist of a hydrophilic sugar alcohol
and a hydrophobic fatty acyl moiety. The diversity of properties
and functions of sugar alcohol fatty acid esters depends largely
on the chemical properties of their building blocks.15–17 In
turn, by the judicious selection of a polar head group with
varying moiety volume, number of hydroxyl groups and fatty
acyl chain length, sugar alcohol fatty acid ester molecules can
be designed for multipurpose applications as surfactants, clea-
ners, stabilizers, emulsifiers, etc. As a result, structurally vari-
able sugar alcohol fatty acid esters may find a variety of
applications in the food, pharmaceutical and cosmetic
industries.7,8,13,18
aZhejiang Key Laboratory for Agro-Food Processing, College of Biosystems
Engineering and Food Science, Zhejiang University, Hangzhou 310058, China
bDepartment of Engineering, Faculty of Science and Technology, Aarhus University,
8000 Aarhus, Denmark. E-mail: guo@eng.au.dk, guo@mb.au.dk;
Fax: (+45) 76123178; Tel: (+45) 89425285
cInterdisciplinary Nanoscience Center, Aarhus University, Gustav Wieds Vej 14,
8000 Aarhus C, Denmark
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
In current commercial markets sugar alcohol fatty acid Rhizomucor miehei, enzyme activity with 25 BIU (batch inter-
esters generally consist of a hydrophobic tail of 8–18 carbons esterification units) g−1, defined as the amount of enzyme
in order to achieve good water solubility.11,13,18 To investigate required to incorporate 1 μmol palmitic acid per min into trio-
the potential value of increased side chain length, we designed leoylglycerol from an equimolar mixture at 40 °C; and
and synthesized a series of sugar alcohol ultra-long fatty acid Novozym 435, immobilized lipase from Candida antarctica
esters with behenic acid (22:0) as a representative. Behenic lipase B, enzyme activity 10 000 PLU (propyl laurate unit) g−1
,
acid was selected not only because it is rarely investigated as a one propyl laurate unit is defined as the number of μmol of
synthetic building block, but also thanks to its ultra-long n-propyl laurate obtained in the standard test corresponding
alkane chain that may enable strong acyl–acyl interactions, to the esterification of lauric acid with n-propyl alcohol after
possibly accounting for unique phase behavior.19,20 Behenic 15 min at atmospheric pressure. Ethylene glycol (99.8%), gly-
acid is known as one of the dominant ultra-long fatty acyls in cerol (≥99%), pentaerythritol (≥99%), xylitol (≥99%), L-(−)-ara-
stratum corneum lipids. It exists as a segment of cera- bitol (≥98%), sorbitol (99%), pancreatin from porcine
mides,21,22 and this is one of the decisive reasons for the pancreas (4 × USP specifications), lipase immobilized on
orthorhombic packing of stratum corneum lipids and the for- Immobead 150 (Pseudomonas fluorescens lipase, ≥600 U g−1),
mation and maintenance of the epidermal water permeability bile salts (cholic acid sodium salt ∼50%, deoxycholic acid
barrier.23–25 Behenic acid has a melting point of 80 °C, hence sodium salt ∼50%), DL-α-palmitin (≥99%), and other chemicals
it can be expected that the resulting compounds from its con- and all solvents used (HPLC grade) were purchased from
jugation with different sugar alcohols will exist as solids. This Sigma-Aldrich (St. Louis, USA). Erythritol (99%) was purchased
new group of compounds might be promising candidates for from VWR (Søborg, Denmark). Phosphatidylcholine (soybean
use as excipients for the manufacture of solid lipid nano- lecithin) was purchased from Epikuron 200, Minneapolis, USA.
particles (SLNs),26–28 as Compritol® 888 (a mixture of behenic Deionized water was generated from a Milli-Q (Millipore, MA)
glycerides) has been a frequently used element in the formu- system. All chemicals were used as received.
lation of SLN delivery systems.28,29 The hydrophilicities and
surface activities of sugar alcohol fatty acid ester molecules are
largely governed by the size and functional groups of the sugar
moiety. To identify a broad spectrum of molecules for multi-
Methods
Preparation and purification of behenic acid
purpose applications, and more importantly to systemically
map the structure–property–function relationship, the sugar
alcohol was altered from ethylene glycol to glycerol, erythritol,
pentaerythritol, arabitol, xylitol and sorbitol. The synthesis of
this series of ultra-long-chain fatty acid sugar alcohol esters
was structurally confirmed by NMR, FT-IR and MS analysis.
The physical properties of the synthetic compounds were pri-
marily characterized by DSC analysis. The packing patterns
and thermostability were investigated by Temp-Ramp-FT-IR. As
a pre-condition for in vivo applications, the ability of the
SAMBs to undergo lipolysis was examined by porcine pancrea-
tic lipase (PPL)-catalyzed hydrolysis. Potential multipurpose
applications, from use as orthorhombic packing enhancers to
use as excipients for the manufacture of surfactant-free nano-
particles, have been identified and assigned to the corres-
ponding structures.
Behenic acid was obtained by the saponification of commer-
cial Compritol® 888 ATO, according to the method described
by Haagsma et al.,30 with some modifications. Compritol® 888
ATO consists of several high melting point long-chain satu-
rated fatty acids (Table 1). The saponification in ethanol was
conducted in a heating flask equipped with a condenser with
circulating water at 5 °C, under reflux conditions.
The purification of behenic acid from the free fatty acid
mixture of saponified Compritol® 888 ATO was carried out by
solvent recrystallization in petroleum ether (boiling point of
Table 1 The fatty acid composition of samples before and after first
and second recrystallizationsa
Fatty acid
composition
Starting
material
First
recrystallization
Second
recrystallization
Materials
C12:0
C14:0
C14:1
C16:0
C18:0
C18:1
C20:0
C20:2
C20:4
C22:0
C22:1
C22:2
C24:0
0.10 0.00
0.05 0.00
0.31 0.02
0.73 0.08
3.22 0.14
0.18 0.00
4.70 0.15
0.13 0.01
0.47 0.13
88.34 0.52
0.10 0.01
0.08 0.00
1.77 0.00
ND
ND
ND
ND
ND
ND
ND
0.24 0.10
ND
2.14 0.08
ND
ND
96.06 0.12
ND
Compritol® 888 ATO was kindly provided as a gift by
BIONORD A/S (Copenhagen, Denmark). Lipozyme TL IM, Lipo-
zyme RM IM and Novozym 435 were obtained from Novozymes
A/S (Bagsvaerd, Denmark). The specifications of the respective
enzymes are as follows: Lipozyme TL IM, a silica granulated
Thermomyces lanuginosa lipase, enzyme activity 175 IUN
(interesterification units Novo) g−1, defined as the amount of
enzyme that converts 0.01% (w/w) of tristearin per min at 70 °C
with a soybean oil substrate (fully hydrogenated soybean oil,
73 : 27, w/w); Lipozyme RM IM, an immobilized lipase from
0.13 0.03
0.88 0.15
ND
2.91 0.23
ND
ND
94.44 0.29
ND
ND
1.66 0.12
ND
1.56 0.10
a ND: not detected.
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
60–80 °C). Typically, 500 mL petroleum ether was added to a
Novozym 435, Lipozyme RM IM, Lipozyme TL IM and Pseu-
glass vial containing 50 g of free fatty acids. This mixture was domonas fluorescens lipase were screened for the potential syn-
stirred at 60 °C until all the free fatty acids were dissolved, and thesis of erythritol monobehenate. Erythritol was selected as a
then kept stirring for another 1 h. The solution was allowed to representative of sugar alcohols for parameter study because it
cool to 30 °C, and then kept at 30 °C for 20 h. The mixture was has two primary and two secondary alcohol groups and exists
filtered under vacuum using an Ace Buchner funnel as a solid at room temperature. Its molecular structure and
(25–50 μm) to obtain the solid (behenic acid-rich fraction) and physical state are also representative of sugar alcohols such as
liquid fractions.
arabitol, xylitol and sorbitol. Because erythritol has a relatively
The fatty acid compositions of the solid and liquid fractions faster reaction rate than arabitol, xylitol and sorbitol in initial
were analyzed by gas chromatography (GC) after methylation, experiments, we used this compound as a test study.
according to the AOAC method.31 The analysis was carried out
Reusability of biocatalyst. A reusability study of Novozym
on a Thermo Trace GC system, equipped with a flame ioniza- 435 was carried with consecutive batch reactions with 2 mmol
tion detector (FID). A Supelco capillary column (Omegawax, behenic acid, 20 mmol erythritol, 50 mg Novozym 435 and
30 m × 0.25 mm) was used. The temperature was programmed 200 mg molecular sieves in 10 mL t-BuOH. The reaction was
to increase from 100 °C to 200 °C at a rate of 60 °C min−1
,
conducted at 60 °C for 12 h, with magnetic agitation at
then to 240 °C at a rate of 5 °C min−1, and then to remain at 350 rpm. At the end of the reaction, the reaction mixture was
this temperature for 10 min. Both the injector and detector subjected to centrifugation and the liquid phase was harvested
were set at 260 °C. Helium was used as the carrier gas, with a by decanting. The conversion of behenic acid was estimated by
flow rate of 1 mL min−1. Fatty acid methyl esters (FAMEs) were analysis of the liquid phase. The recovered enzyme, molecular
identified by comparison of the retention times with fatty acid sieves and unreacted erythritol were used for the next batch
methyl ester standard 17AA (Nu-Chek Prep, MN, USA). Area reaction with fresh behenic acid in t-BuOH, with the addition
percentage was used as weight to estimate the relative concen- of a certain amount of erythritol to compensate for the con-
trations of the FAMEs. The determinations of all samples were sumption during the reaction. The Novozym 435 reusability
carried out in triplicate.
experiment was done under mild conditions avoiding strong
physical damage (no vigorous stirring, centrifuge, filtering or
washing) and exposure to air for a prolonged period. This
General procedure for lipase-catalyzed synthesis
A typical reaction mixture consists of 2 mmol behenic acid, process was repeated 14 times.
20–30 mmol sugar alcohol, 200 mg molecular sieves (3 Å, acti-
Purification of synthetic compounds. The reaction mixture
vated by heating up to 250 °C for 8 h) and t-BuOH. The reac- containing compound 1, 2 or 3a was dissolved in 50 mL DCM,
tion mixture was incubated in a 50 mL jacketed glass reactor washed three times with saturated Na2CO3 solution to remove
with the temperature controlled at 60 °C by a circulating water the unreacted behenic acid, and then three times with satu-
bath. After being heated at 60 °C for 30 min, the reaction was rated NaCl solution to completely remove excessive ethylene
initiated by the addition of 5%–8% Novozym 435 (based on glycol, glycerol or erythritol. After removal of the trace
behenic acid weight), with magnetic agitation at 350 rpm. amounts of water with anhydrous sodium sulfate, the organic
Thin layer chromatography (TLC) analysis was used to monitor phase was separated and the solvent was removed by rotatory
the reaction progress.2 Aliquots of samples from the reaction evaporation under vacuum. For the purification of compounds
mixture were taken at set time intervals, diluted with 300 μL 3b, 4a or 4b, the unreacted behenic acid in the reaction
CHCl3/methanol 3 : 1 (v/v) and analyzed on TLC plates (TLC mixture was removed by washing with n-hexane. 50 mL
silica gel 60, 5 cm × 10 cm, Merck, Germany). The plates were n-hexane was added to the reaction mixture and the mixture
developed twice, first with I: CHCl3/methanol/water 64 : 10 : 1 was stirred at 50 °C for 10 min. Then the mixture was filtered
(v/v/v) until the solvent frontier moved 6 cm. After completely and the resulting solid was washed two more times with
drying, the plates were then developed with II: CHCl3/metha- hexane to remove the behenic acid completely. The unreacted
nol/acetic acid 97.5 : 2.5 : 1 (v/v/v) until the solvent front moved sugar alcohol in the obtained solid was removed by washing
8 cm. The plates were then dried again and sprayed with 30% with 50 mL Milli Q water at room temperature three times. The
sulfuric acid in methanol. At the end of the reaction, the lipase resulting solid product was dried in the oven at 60 °C over-
and molecular sieves were removed by filtration and the night. In the case of compound 5, the behenic acid was
solvent was removed by rotatory evaporation under vacuum. removed by washing with 20 mL DCM three times. The
The conversion of behenic acid was estimated by thin layer remaining purification procedure was the same as for com-
chromatography with flame ionization detector (TLC-FID) ana- pound 3b. Identification of the compounds was carried out by
lysis.32 Briefly, the sample was dissolved in CHCl3/methanol 1H NMR spectroscopy analysis (Bruker Avance III spectro-
3 : 1 (v/v) (10 mg mL−1) and a 1.2 μL aliquot was spotted and meter) at 400 MHz. MS spectra were obtained on an instru-
developed in CHCl3/methanol/water 64 : 10 : 1 (v/v/v). The ment operating with electrospray ionization and ion-trap
peaks were identified with reference to the external standards. (ESI-IT) quadrupole detection (Bruker Daltonic GmbH,
Area percentage was used as weight for the calculation of the Bremen, Germany).
yields of the products. The determination was performed in
duplicate.
Synthesis of 1-O-docosanoylethylene glycol (ethylene glycol–
behenic acid monoester, compound 1). 2 mmol of behenic
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
acid and 24 mmol of ethylene glycol were catalyzed by total weight of equivalent amount of behenic acid) in 10 mL
1
Novozym 435 (6% of total weight of equivalent amount of t-BuOH for 24 h. White solid; Rf 0.31; m.p. 83.33 °C; H NMR
behenic acid) in 5 mL t-BuOH for 12 h. White solid; Rf 0.69; m. (400 MHz, CDCl3, 25 °C, TMS): δ = 4.15–4.06 (m, 2H, –CH2–O–
p. 70.59 °C; 1H NMR (400 MHz, CDCl3, 25 °C, TMS): δ = CvO), 3.90–3.84 (m, 1H, –CH–OH), 3.70–3.64 (m, 1H, –CH–
4.25–4.18 (m, 2H, –CH2–O–CvO), 3.87–3.79 (m, 2H, –CH2– OH), 3.62–3.55 (m, 2H, –CH2–OH), 3.53 (t, J = 3.5 Hz, 1H,
OH), 2.32 (t, J = 7.6 Hz, 2H, –CH2–CvO), 2.15 (bs, 1H, –OH), –CH–OH), 2.26 (t, J = 7.6 Hz, 2H, –CH2–CvO), 1.65–1.45
1.73–1.54 (m, 2H, –CH2–), 1.36–1.18 (m, 36H, –CH2–), 0.87 (m, 2H, –CH2–), 1.42–0.94 (m, 36H, –CH2–), 0.79 (t, J = 6.8 Hz,
(t, J = 6.7 Hz, 3H, –CH3); MS, m/z calcd for C24H48O: 384.360; 3H, –CH3); MS, m/z calcd for C27H54O6: 474.392; found:
found: 407.3483 (M + Na+).
497.3785 (M + Na+).
Synthesis of 1-O-docosanoylglycerol (glycerol–behenic acid
Synthesis of 1-O-docosanoylsorbitol (sorbitol–behenic acid
monoester, compound 2). 2 mmol of behenic acid and monoester, compound 5). 2 mmol of behenic acid and
24 mmol of glycerol were catalyzed by Novozym 435 (8% of 20 mmol of sorbitol were catalyzed by Novozym 435 (20% of
total weight of equivalent amount of behenic acid) in 5 mL total weight of equivalent amount of behenic acid) in 10 mL
1
t-BuOH for 12 h. White solid; Rf 0.43; m.p. 83.57 °C; H NMR t-BuOH for 24 h. White solid; Rf 0.21; m.p. 110.29 °C; 1H NMR
(400 MHz, CDCl3, 25 °C, TMS): δ = 4.30–4.05 (m, 2H, –CH2–O– (400 MHz, CDCl3, 25 °C, TMS): δ = 4.34–4.05 (m, 2H, –CH2–O–
CvO), 3.96–3.91 (m, 1H, –CH–OH), 3.74–3.56 (m, 2H, –CH2– CvO), 3.80 (d, J = 5.3 Hz, 1H, –CH–OH), 3.76–3.67 (m, 2H,
OH), 2.78 (bs, 2H, 2 × –OH), 2.35 (t, J = 7.6 Hz, 2H, –CH2–), –CH2–OH), 3.67–3.52 (m, 3H, –CH–OH), 2.29 (t, 2H, –CH2–
1.68–1.55 (m, 2H, –CH2–), 1.37–1.15 (m, 36H, –CH2–), 0.87 CvO), 1.64–1.47 (m, 2H, –CH2–), 1.42–0.98 (m, 36H, –CH2–),
(t, J = 6.7 Hz, 3H, –CH3); MS, m/z calcd for C25H50O4: 414.371; 0.80 (t, J = 6.6 Hz, 3H, –CH3); MS, m/z calcd for C27H54O6:
found: 437.3576 (M + Na+).
504.403; found: 505.4091 (M + H+).
Synthesis of 1-O-docosanoylerythritol (erythritol–behenic
acid monoester, compound 3a). 2 mmol of behenic acid and
Differential scanning calorimetry
20 mmol of erythritol were catalyzed by Novozym 435 (15% of The thermal properties were analyzed using differential scan-
total weight of equivalent amount of behenic acid) in 10 mL ning calorimetry on a Pyris 6 DSC system (Perkin-Elmer Cetus,
1
t-BuOH for 24 h. White solid; Rf 0.34; m.p. 85.49 °C; H NMR Norwalk, USA). Approximately 8 mg of each sample was put
(400 MHz, CDCl3, 25 °C, TMS): δ = 4.25–4.06 (m, 2H, –CH2–O– into an aluminum pan and placed in the equipment under a
CvO), 3.74–3.67 (m, 1H, –CH–OH), 3.67–3.57 (m, 2H, –CH2– purging atmosphere of nitrogen (20 mL min−1), with an empty
OH), 3.54–3.46 (m, 1H, –CH–OH), 2.28 (t, J = 7.6 Hz, 2H, pan as an inert reference. The heating and cooling profile was:
–CH2–CvO), 1.60–1.47 (m, 2H, –CH2–), 1.29–1.09 (m, 36H, (1) initial temperature 20 °C; (2) ramp 20 K min−1 to 120 °C;
–CH2–), 0.80 (t, J = 6.8 Hz, 3H, –CH3); MS, m/z calcd for (3) isothermal for 5 min; (4) ramp −5 K min−1 to 60 °C; (5) iso-
C26H52O5: 444.381; found: 467.3696 (M + Na+).
thermal for 10 min; and (6) ramp 5 K min−1 to 120 °C. The
Synthesis of 1-O-docosanoylpentaerythritol (pentaerythritol– DSC scans were evaluated using MicroCal Origin 8.6 software.
behenic acid monoester, compound 3b). 2 mmol of behenic
Fourier transform infrared spectroscopy
acid and 30 mmol of pentaerythritol were catalyzed by
Novozym 435 (15% of total weight of equivalent amount of Fourier transform infrared spectroscopy (FT-IR) was used for
behenic acid) in 10 mL t-BuOH for 48 h. White solid; Rf 0.37; the detection and characterization of the lipid organization in
m.p. 78.37 °C; 1H NMR (400 MHz, CDCl3, 25 °C, TMS): δ = 4.15 the synthetic compounds. Samples were pressed onto a ZnSe
(s, 2H, –CH2–O–CvO), 3.58 (s, 6H, –CH2–OH), 2.29 (t, J = ATR crystal mounted in a trough plate and placed in a horizon-
7.6 Hz, 2H, –CH2–CvO), 1.64–1.49 (m, 2H, –CH2–), 1.29–1.10 tal ATR-FT-IR spectrometer (PIKE, Madison, WI; Bruker, Ettlin-
(m, 36H, –CH2–), 0.81 (t, J = 6.8 Hz, 3H, –CH3); MS, m/z calcd gen, Germany). The temperature of the ATR plate with loaded
for C27H54O5: 458.397; found: 459.4019 (M + H+).
sample was precisely controlled and monitored by a tempera-
Synthesis of 1-O-docosanoylarabitol (arabitol–behenic acid ture controller. Spectra in the ATR setup were acquired with
monoester, compound 4a). 2 mmol of behenic acid and 2 °C intervals from 40–80 °C. The spectra were collected with
20 mmol of arabitol were catalyzed by Novozym 435 (20% of an unpolarized beam at a resolution of 4 cm−1 with 4 scans,
total weight of equivalent amount of behenic acid) in 10 mL and a spectral window of 4000–600 cm−1. Background spectra
1
t-BuOH for 24 h. White solid; Rf 0.30; m.p. 86.71 °C; H NMR of the clean ZnSe crystal disc were collected at initial tempera-
(400 MHz, CDCl3, 25 °C, TMS): δ = 4.15–4.06 (m, 2H, –CH2–O– tures and subtracted from the sample spectra. The FT-IR
CvO), 3.90–3.84 (m, 1H, –CH–OH), 3.70–3.64 (m, 1H, –CH– spectra were analyzed using MicroCal Origin 8.6 software.
OH), 3.62–3.55 (m, 2H, –CH2–OH), 3.53 (t, J = 3.5 Hz, 1H,
Critical micelle concentration
–CH–OH), 2.26 (t, J = 7.6 Hz, 2H, –CH2–CvO), 1.65–1.45
(m, 2H, –CH2–), 1.42–0.94 (m, 36H, –CH2–), 0.79 (t, J = 6.8 Hz, The critical micelle concentrations (CMCs) of the different
3H, –CH3); MS, m/z calcd for C27H54O6: 474.392; found: sugar alcohol behenic acid monoesters were determined with
497.3801 (M + Na+).
pyrene fluorescence33 using
a fluorescence spectrometer
Synthesis of 1-O-docosanoylxylitol (xylitol–behenic acid (Varian Cary Eclipse, Agilent Technology, California, USA).
monoester, compound 4b). 2 mmol of behenic acid and Sample solutions of different concentrations (1, 0.5, 0.1, 0.05,
20 mmol of xylitol were catalyzed by Novozym 435 (20% of 0.01, 0.005 and 0.001 mg mL−1) were prepared using water
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
previously saturated with pyrene (final concentration of 1 μM). an in-lens secondary electron detector with an accelerating
Emission spectra were obtained by exciting the samples at voltage of 5 kV and a working distance of 4–5 mm.
343 nm. The fluorescence intensity ratio of I1/I3 was plotted
against sample solution concentration.
Results and discussion
Porcine pancreatic lipase-catalyzed hydrolysis of synthetic
Preparation and purification of benenic acid
sugar alcohol monobehenates to examine lipolysis stability
To obtain high-purity behenic acid at a relatively cheap price,
we established a method to prepare and purify this ultra-long
Evaluation of the in vitro digestion profiles of the synthetic
sugar alcohol behenic acid monoesters was conducted by
fatty acid from Compritol® 888 ATO, which is a commercially
porcine pancreatic lipase (PPL)-catalyzed hydrolysis compared
available product of a glyceride mixture rich in behenic acid.
to that with glycerol palmitic acid monoester, according to the
Compritol® 888 ATO consists of 18% mono-, 52% di- and 28%
method of Sek et al.,34,35 with some modification. Standard
tribehenate of glycerol (by wt%), and 88% of the constituted
digestion buffer (Tris-HCl, pH = 7.50, 50 mM; NaCl, 150 mM;
CaCl2, 5 mM), a PC/bile salts mixed micellar solution (PC,
5 mM; bile salts, 20 mM) in digestion buffer and pancreatin
extract (1 g in 5 mL digestion buffer), was prepared and stored
fatty acids is behenic acid.37,38 The saponification was carried
out in ethanol under reflux conditions (>80 °C) to assure all
components in Compritol® 888 melted, and up to 99.5% con-
version was achieved. In the obtained free fatty acid mixture
(Table 1), apart from behenic acid there were some other high-
at 4 °C. Briefly, 25 mg of each sugar alcohol behenic acid
monoester was dispersed by emulsification in 1 mL of PC/bile
melting-point fatty acids such as lignoceric acid (24:0), which
salts mixed micellar solution at 37 °C with continuous stirring
has low solubility in low boiling solvents like acetone
(b.p. 56 °C). We thus chose petroleum ether (b.p. range of
60–80 °C) as the solvent for the purification of behenic acid.
According to the fatty acid composition analysis results shown
in Table 1, all the unsaturated and medium-chain fatty acids
were removed by the first recrystallization. The behenic acid
was enriched to 94% from 88%. The second recrystallization
at 200 rpm for 10 min. Then, 100 μL of pancreatin extract was
added to initiate lipolysis. The lipolysis experiments were
allowed to continue for 60 min. During the reaction, 20 μL ali-
quots of the samples were collected and acidified with 10 μL of
1 M HCl and 200 μL CHCl3/methanol 2 : 1 (v/v) in EP tubes for
0, 1, 2, 5, 10, 30 and 60 min. After centrifugation at 4000 rpm
for 15 min at 25 °C, the suspensions were separated and ana-
resulted in an increase of behenic acid content to 96%, with
lyzed by TLC-FID, using the same method as described above.
the overall yield of 78%. As the other saturated long-chain fatty
acids such as arachidic acid (20:0) and lignoceric acid have
Preparation of solid lipid nanoparticles (SLNs) from synthetic
sugar alcohol monobehenates
similar solubilizing behavior in petroleum ether to behenic
acid, they cannot be removed completely by recrystallization.
Further recrystallization was not carried out in this work
because the recovery yield decreased remarkably, whilst the
behenic acid content only increased marginally. The obtained
product appears as white crystals with a melting point of
81.97 °C (expected m.p. 80 °C).
SLNs were prepared with an emulsification–diffusion approach
based on a previously reported method, with slight modifi-
cation.36 Briefly, 15 mg samples were dissolved in 1.6 mL
water saturated with methyl ethyl ketone in a water bath at
65 °C. Then 4 mL methyl ethyl ketone-saturated water was
added, and the mixture was stirred with a mechanical stirrer at
1500 rpm for 3 min to form O/W emulsions. Afterwards,
16 mL Milli Q water was added and the mixture was stirred at
Lipase-catalyzed synthesis of sugar alcohol behenic acid
monoesters
300 rpm for 5 min, in order to cause solvent diffusion to the There are a few factors governing the determination of an enzy-
continuous phase and to initiate lipid aggregation and the for- matic synthesis strategy: selection of solvent system, compat-
mation of nanoparticles.
ibility with enzyme and reaction selectivity. In the present
The sizes of the SLNs were measured with a Malvern Zetasi- work, the target molecules, ultra-long fatty acid sugar alcohol
zer Nano (Malvern, Worcestershire, UK). The sample solutions monoesters, are presumed to be obtained by coupling an ultra-
were made by adding 0.1 mL of fresh SLNs into 0.8 mL of Milli long-chain fatty acid with a sugar alcohol containing multiple
Q Water, and were analyzed in triplicate with a reference hydroxyl groups, which represent two extreme cases in terms
refractive index of 1.553. Morphology studies were carried out of hydrophilic and hydrophobic properties. The ultra-long-
using scanning electron microscopy (SEM) Nova NanoSEM 600 chain fatty acids such as behenic acid are strongly hydro-
(FEI, Eindhoven, The Netherlands). The silicon substrates were phobic, while sugar alcohols are highly hydrophilic, hence
cleaned by sonication in ethanol for 5 min and water for choosing a compatible reaction system is important. Strong
5 min, and then dried with a nitrogen stream. 5 μL of fresh polar solvents such as DMSO are not a good choice because
SLN solutions were added on the substrates, and stuck on to a enzyme will be deactivated by their strong solvation effect. To
holder through electrically conductive silicone sheets. The cope with this challenge, some approaches have been
holder was then kept in a vacuum desiccator overnight for attempted, such as solvent-free biosynthesis at elevated tem-
drying of the samples. Dehydrated nanoparticles were coated peratures under vacuum,39 and using property-tunable ionic
with gold. The morphology of the particles was observed using liquid (IL) systems.40,41 The reaction at high temperature
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
under vacuum is not appropriate for multi-OH sugar alcohols,
as it leads to low selectivity and increasing formation of side
products. Ionic liquids might be a good system, however, the
cost, compatibility of ILs with enzymes and post-reaction sep-
aration and purification of the products are the factors that
should be taken into consideration.42–44
In the present case, t-BuOH, benign to enzymes, was
selected as a solvent in which both behenic acid and sugar
alcohols have considerable solubility. Excessive sugar alcohol
was employed to reduce the probability of formation of multi-
substituted sugar alcohols and to drive the shifting of the reac-
tion equilibrium to the right. Molecular sieves were added to
remove the water generated from esterification, thus contribut-
ing to the progress of the forward reaction.40–44 The influences
of lipase type, temperature and time on the biosynthesis of
compound 3a are presented in Fig. 1. Screening of lipases was
performed with the esterification of 2 mmol behenic acid with
20 mmol erythritol in 10 mL t-BuOH, and 15% Novozym 435
(by the total weight of behenic acid and erythritol) gave the
highest conversion (53%) among the test lipases (Fig. 1(A)).
The increase in reaction temperature from 40 to 60 °C resulted
in a marked increase in the conversion of behenic acid (21%
to 53%), however, further increase in temperature did not lead
to a corresponding enhancement of conversion (Fig. 1(B)). The
reaction time course study demonstrated that equilibrium is
essentially attained in 24–30 h (Fig. 1(C)).
As shown in Fig. 1(D), after 15 batches of consecutive reac-
tions, no significant decrease of conversion of behenic acid
was observed (reaction time of 180 h). Conversions of behenic
acid varied in the range of 47–51%, indicating excellent reusa-
bility of Novozym 435 in the synthesis of erythritol monobehe-
nate, most likely due to the low operation temperature and
protective effect from t-BuOH,45,46 indicating the commercial
value of the process developed in this work.
The optimal conditions established by single parameter
study for the synthesis of the compound 3a were further
extended to the esterification of behenic acid with a variety of
sugar alcohols with slight adjustment for individual reactions,
and the results are summarized in Table 2. Among all the
examined sugar alcohols, only ethylene glycol and glycerol are
completely miscible with t-BuOH. The rest of the sugar alco-
hols are only partially soluble, therefore, in the reaction system
the presence of solid sugar alcohol was used to indicate that
the reaction system was saturated. A decreasing yield was
obtained for multi-OH sugar alcohols because of their decreas-
ing solubility in t-butanol. Thus, higher enzyme loading
and longer reaction times were needed for multi-OH sugar
alcohols in order to achieve considerable conversion. The reac-
tion progress was monitored by periodic sampling and TLC
analysis (Fig. 2), and the reaction was terminated when a con-
siderable conversion was achieved while little sugar alcohol
behenic acid diester was formed (Table 2). The molecular
structure of pentaerythritol is different from other sugar
alcohols. As it has a pyramidal grouping structure, with
four CH2OH groups around the central carbon atom in place
Fig. 1 Effects of lipases (A), temperature (B) and reaction time (C) on
the lipase-catalyzed synthesis of compound 3a, as well as the reusability
of Novozym 435 in consecutive batch reactions (D). See the details for
the reaction conditions in Methods, section 2, General procedure for
lipase-catalyzed synthesis.
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
Table 2 Reaction scheme and results of the lipase-catalyzed synthesis of sugar alcohol behenic acid monoesters
t-BuOH
(mL)
Time
(h)
Lipase
(%)
Conversion
(%)
Isolated
yield (%)
Compounds
Ratio
1
2
3a
3b
4a
4b
5
12
12
10
15
10
10
10
5
5
10
10
10
10
10
12
12
24
48
24
24
24
6
8
15
15
20
20
20
90
79
61
30
53
46
36
57
37
26
24
22
21
23
Fig. 2 Molecular structures of the synthetic compounds and their assay on a TLC plate.
of the generally accepted tetrahedral distribution,47 it has Synthesis of 1-O-docosanoylpentaerythritol (pentaerythritol–
strong intramolecular networking. Therefore, the highest behenic acid monoester, compound 3b).
alcohol ratio and longest reaction time were required to
The purification processes were treated differently for indi-
achieve relatively high conversion (see the Methods section, vidual synthetic compounds (Fig. 2). Compounds 1, 2 and 3a
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
have relatively small hydrophilic heads compared to the long-
chain fatty acid hydrophobic tail. They are soluble in DCM/
CHCl3. The unreacted behenic acid could be removed by
washing with a Na2CO3 solution to form behenic acid sodium
salt. The unreacted sugar alcohol partitions into the aqueous
phase. For compound 3a, it has more amphiphilic properties,
and tends to stay at the interface of the two phases, thus result-
ing in a large loss during the purification process. While com-
pounds 3b, 4a, 4b and 5 have even stronger amphiphilic
properties, and are difficult to dissolve in both polar and non-
polar solvents. Therefore, their purifications were carried out
by washing with a non-polar solvent (n-hexane) to remove the
unreacted behenic acid, and the sugar alcohols were washed
out with deionized water.
The obtained synthetic compounds 1, 2, 3a, 4a, 4b and 5
1
were subjected to structural identification by H NMR and MS
analysis. As expected the monoester is the dominating
product. This has been detailed in the Methods section. It
should be pointed out that, even though the enzymatic reac-
tion is highly selective and the dominant product is the sugar
alcohol monoester (Fig. 2), in the syntheses of compounds 2,
4a and 4b the behenic acyl is also conjugated to a secondary
alcohol, as observed in the DSC thermogram curves (Fig. 3).
Fig. 3 DSC thermograms of sugar alcohol monobehenates.
the common behenyl endows a certain amount of similarity
on their physicochemical properties, while the difference in
properties is largely determined by the differing sugar alcohol
moieties. This makes it possible to create a broad spectrum of
molecules that could be assigned to different applications.
Moreover, the cooperation between hydrophilic and hydro-
phobic moieties is always a key factor to be considered when
one analyzes the physicochemical properties of the
compounds.
DSC thermogram. Differential scanning calorimetry is one
of the most frequently used techniques to determine the
thermal properties of solid materials.48 The intra- and inter-
molecular interactions during the thermal process can be
reflected from characteristic thermal parameters. The thermal
behaviors of the synthetic compounds have thus been charac-
terized by DSC in terms of thermal temperature and the
enthalpy of the phase transition, and the results are summar-
ized in Table 3.
Physicochemical characteristics of sugar alcohol behenic acid
monoesters
To the best of our knowledge, this work presents the first sys-
temic study of an array of ultra-long fatty acid sugar alcohol
esters with altering polar heads from ethylene glycol to sorbi-
tol. Therefore, characterization of their physicochemical pro-
perties (such as thermal properties, molecular packing
behavior, lipolysis stability and surface-active properties) is of
fundamental interest. Most importantly, an understanding of
the molecular basis accounting for their respective properties
and their correlation is the first step to identify the areas of
potential applications. Inspection of the molecular structures
of the synthetic compounds revealed that, in addition to the
common behenic acyl structure, from compound 1 to 5 the
molecular size and volume increases as the numbers of carbon
and OH increase (Fig. 2). The copresence of ultra-long acyl
chain behenyl and polyhydroxyl alcohol endows this group of
compounds with amphiphilic properties. It is anticipated that
A close overview of the thermal data shown in Table 3
shows that: (1) Tm increases in the order of 1 < 3b < 4b ≤ 2 <
3a < 4a < 5, and the change in onset temperature has a similar
Table 3 Thermal properties of synthetic compoundsa
Compounds
M.F.
M.W.
Tonset (°C)
Tm (°C)
ΔT1/2 (°C)
ΔH (J g−1
)
CMC (µM)
1
2
3a
3b
4a
4b
5
C
24H48O3
384.64
414.66
444.69
458.71
474.71
474.71
504.74
66.69
79.90
80.28
70.72
82.30
77.24
105.16
70.59
83.57
85.49
78.37
86.71
83.33
110.29
3.56
5.89
8.87
7.21
9.33
10.85
10.74
190.11
138.17
73.56
108.36
60.59
58.59
39.03
15.4
89.9
90.0
64.0
111.8
90.9
C25H50O4
C26H52O5
C27H54O6
C27H54O6
C27H54O6
C28H56O7
157.3
a M.F., molecular formula; M.W., molecular weight; Tonset: temperature at which the thermal effect starts; Tm: temperature at which heat capacity
at constant pressure is maximum; ΔT1/2: temperature width at which the transition is half completed; ΔH: transition enthalpy.
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
trend; (2) the temperature width (ΔT1/2) at which the transition the Tonset, the phase of the compound starts transforming into
is half completed decreases in the order of 1 < 2 < 3b < 3a < 4a liquid crystalline. A larger peak width at half maximum indi-
< 4b < 5; and (3) the transition enthalpy ΔH has the same cates a broader gel-to-liquid crystalline transition.49 Interest-
decreasing order as ΔT1/2. These results are interesting and ingly, ΔT1/2 increases as M.W. increases, in exactly the same
deserve further analysis. Clearly, one can say that the melting order as ΔH decreases (Table 3). A larger ΔT1/2 value is a
points of the synthetic compounds increase, in general, as the characteristic indication of the slow breaking down of the
molecular weight or number of OH units in the sugar alcohol H-bonding network. A logical explanation is that the strong
moiety increases, but it does not exactly follow the same order. intermolecular H-bonding is a natural consequence of the
Definitely this changing trend could be correlated to the intro- presence of a sugar alcohol with multiple OH groups (Fig. 2).
duced sugar alcohol structure, where the OH groups could Thus, we can postulate that the melting process of the sugar
enable intra- and intermolecular hydrogen bonding. For alcohol behenate begins with a thermodynamic transition of
example, erythritol in compound 3a and pentaerythritol in acyl chain packing from an all-trans conformation to increas-
compound 3b have the same number of OH units, however, ing gauche conformation, with simultaneous disassociation of
pentaerythritol structurally favors the formation of intra- the intermolecular network formed by the polar heads,47,48
molecular H-bonding and erythritol favors intermolecular until the transition from solid gel to liquid crystalline is com-
H-bonding (intermolecular association), which could explain pleted.48 Therefore, the broad ΔT1/2 is the cooperative result of
why compound 3a has a higher m.p. (Table 3). However, van der Waals and H-bond interactions. This hypothesis has
H-bonding from the polar head is not the only factor to deter- been verified by FT-IR analysis, as presented in the following
mine the change in m.p. of the synthetic compounds; the role section.
of van der Waals interactions between behenyl chains and the
The large sugar alcohol head group and intermolecular
cooperation between behenyl and sugar alcohol segments H-bond networking could decrease head group mobility and
cannot be neglected. For instance, xylitol in compound 4b affect the alkyl chain packing. This molecular feature may
(having four OH groups) and glycerol in compound 2 (having enable intermolecular association and arrangement for supra-
two OH groups) have almost the same m.p. value. A possible molecular structures, such as micelles. Thanks to the high
explanation is that, the greater number of OH groups in com- melting points of the synthetic compounds in this work, this
pound 4b increase the number of H-bonding interactions, but array of compounds might be applied as excipients for the
the increasing volume of polar heads in compound 4b also preparation of SLNs. In this respect, the structure and pro-
perturbs the intermolecular packing of the behenyl chains and perties of the sugar alcohol polar heads will influence the mor-
thus weakens the van der Waals interactions. Overall, all of the phology structure of the SLNs and the release of encapsulated
synthetic compounds have higher m.p. values than their drugs.50,51
parent compound behenic acid, and this new array of mole-
cules may provide the possibility for innovative applications.
FT-IR spectroscopy. Although a lot of state-of-the-art tech-
niques have been established to study molecular structure and
As presented in Table 3, ΔH (which refers to the energy organization, FT-IR is still a quite powerful, rapid, inexpensive
added to turn the compound from a solid to a liquid) shows a and nonperturbing technique for studying lipid phase tran-
distinctly different trend from the change in m.p., namely, it sitions.52,53 DSC can provide the information of temperature,
decreases as M.W. increases. This seems to indicate that enthalpy and cooperativity during the phase transition,
enhanced intermolecular H-bond interactions resulting from however, no organization information can be obtained. FT-IR
the presence of a multi-OH sugar alcohol moiety do not con- spectroscopy is a potent technique for probing the confor-
tribute to an increase in heating capacity, rather they decrease mations and orientations of the alkyl chain packing. In par-
the thermal stability of the solid lipid. Most likely this is ticular, the frequencies of the methylene symmetric and
because the ΔH value is determined mainly by fatty acyl chain antisymmetric vibrational stretches are known to relate to the
packing mode and pattern. Ethylene glycol behenate (com- population of trans and gauche conformers. In this work, vari-
pound 1) and glycerol behenate (compound 2) have smaller able-temperature transmission FT-IR was used to study the
polar heads that give little perturbance to the packing of fatty hydrocarbon chain packing behavior and phase transitions of
acyl chains. However, with increasing volume of polar head the synthesized sugar alcohol behenates in order to gain more
the intermolecular distance between adjacent fatty acyl chains insight into the chemistry and physics determining their pro-
becomes larger, and the packing becomes less compacted.44,47 perties. It is well known that the functional group or molecular
Hence the intermolecular van der Waals interactions get structure could be assigned to specific FT-IR absorption peaks.
weaker, which corresponds to a decreasing ΔH. This assump- Considering the common structural characteristics in sugar
tion is supported by FT-IR analysis of lipid packing, as detailed alcohol fatty acid esters, we chose three absorption bands to
in the following section.
monitor the acyl chain melting transitions, which are the CH2
Fig. 3 shows the DSC endotherms of the various com- bending band (∼1470 cm−1), the CH2 symmetric stretching
pounds. The results indicate that molecules with different band (2920 and 2850 cm−1) and the CvO stretching band
sizes of sugar alcohol head groups have different peak widths, (∼1700 cm−1).
which is characterized as ΔT1/2, the temperature of transition
Compound 1, having a smaller polar head, and compound
at half height (Table 3). When the temperature is higher than 5, having a larger polar head, were chosen as representatives
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 4 Variable-temperature absorbance FT-IR spectra of the thermotropic phase transitions of compound 1 (A, C, E) and compound 5 (B, D, F).
for FT-IR analysis (Fig. 4). Fig. 4A and B tracks the thermotro- patterns of the rest of the synthetic compounds and their col-
pic response of the CH2 bending mode frequencies of com- lapse temperatures are summarized in Table 4. The data in
pound 1 and compound 5 as a function of temperature. The Table 4 provides direct evidence that compounds 1 and 2 have
CH2 bending mode of compound 1 is split into two bands, highly ordered orthorhombic chain packing at physiological
which is a characteristic of orthorhombic packing. Contrarily temperature. In the orthorhombic packing all acyl methylene
only a single peak is observed for compound 5, indicating groups are assembled in the trans conformation, for which
compound 5 is arranged with hexagonal packing. The packing high energy is needed to break the packing network. This
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
collapse temperatures (Table 4) are significantly lower than the
corresponding onset temperature of melting (Table 3),
suggesting at collapse temperature the interactions between
adjacent molecules are disordered but not completely random-
ized and disassociated.
Table 4 Lipid packing modes of the synthetic compounds and their
collapse temperatures
Compounds
Lipid packing
T (°C)
1
2
3a
3b
4a
4b
5
Orth
Orth
Hex
Hex
Hex
Hex
Hex
48
58
62
54
60
60
66
CMC. The amphiphilic compounds that contain hydro-
phobic as well as hydrophilic moieties tend to form micelles
or bilayers through self-assembly when the concentration is
higher than a certain threshold.56 The minimum concen-
tration to form micelles (critical micelle concentration, CMC)
was measured for the synthetic ultra-long fatty acid sugar
alcohol esters, as summarized in Table 3. The CMC value
depends on intrinsic factors such as the structures of the
hydrophobic and hydrophilic parts of the amphiphilic mole-
cule, and external factors such as medium temperature and
composition (ionic strength, dielectric constant, and pH).57 In
this study, as the synthetic compounds share the same hydro-
phobic behenoyl group, so the difference in hydrophilicity
between these molecules is largely determined by the pro-
perties of the polar head (sugar alcohol moiety).
As depicted in Fig. 6, the CMC value increases in general as
the number of OH groups in the sugar alcohol segment
increases, with compounds 3b and 4b as exceptions, likely due
to their strong intramolecular H-bonding. For example, the
sugar moiety of compound 3b has a tetrahedral structure, and
can form stable intramolecular H-bonding, which restrains its
solubility in water. It should be pointed out that the CMC
values of all the synthetic compounds are in a low range
(15–150 μM), which means they tend to form micelles at very
low concentrations. The main driving force in behenoyl sugar
alcohol esters for the formation of micelles at low CMC values
is their strong hydrophobic tails, resulting from the ultra-long
behenoyl group, which is consistent with Priev’s concept that
the critical micelle concentration decreases exponentially with
amphiphilic chain length.56
result thus explains the remarkably higher transition enthalpy
of compounds 1 and 2 compared to the hexagonal packing
compounds (3b, 4a, 4b and 5) as measured by DSC (Table 3).
The CH2 symmetric stretching and CH2 anti-symmetric
stretching vibrations give rise to IR-active bands centered near
2850 cm−1 and 2920 cm−1, respectively. The thermotropic
responses of the CH2 symmetric stretching band frequencies
of compound 1 and compound 5 are shown in Fig. 4C and D,
respectively. Upon alkyl packing collapse, the conformational
disordering of an all-trans polymethylene chain of compound
1 is accompanied by an upward shift in the CH2 symmetric
stretching and CH2 anti-symmetric stretching band maxima by
3–4 cm−1 and 5–7 cm−1, respectively (Fig. 4C), and by a signifi-
cant broadening of the overall band envelope.54 A similar
changing trend is observed for compound 5. These changes
reflect the increases in hydrocarbon chain conformational dis-
order and mobility that occur with the onset of gauche rotamer
formation and the concomitant decline in the number of all-
trans rotamers present upon lipid melting.
The influences of the headgroups to the intramolecular
chain packing behavior have also been studied (Fig. 4E and F)
by examining the frequencies of the fatty acid carbonyl stretch-
ing band, which is sensitive to the hydrogen bonding.55 When
the temperature increases to the phase transition temperature,
the hydrogen bonding breaks and the head group interactions
are disconnected. This is reflected as the shifting of the fre-
quency. As compound 1 has only one hydroxyl group, there is
a sharp jump when the hydrogen bonding breaks (Fig. 4E).
While for compound 5, which has five hydroxyl groups, the
shifting occurs at higher temperature (Fig. 4F), indicating the
presence of a stronger intermolecular H-bonding network, for
which more energy is needed to dissociate the interactions.
The FT-IR absorption frequency change has also been
further illustrated by plotting the peak position as a function
of temperature (Fig. 5a′–f′). Interestingly, although the discon-
tinuity of the FT-IR thermal spectrum behaves in different
ways (Fig. 4), for the same compound, different FT-IR absorp-
tion bands (Fig. 5a′ 5c′ and 5e′ and Fig. 5b′ 5d′ and 5f′), which
correspond to different functional groups, show the same col-
lapse temperature (48 °C for compound 1 and 66 °C for com-
pound 5). This suggests that at collapse temperature the
molecular packing is completely disordered for both hydro-
carbon chains and polar heads. It is noticed that the packing
It is interesting to see that there is good consistency
between the change in CMC values and the m.p. values of the
respective compounds (Fig. 6). This is probably because the
melting point of the synthetic behenoyl alcohol ester is associ-
ated with the number of OH groups and size of the sugar
alcohol segment. In turn, these values influence the molecular
solvation by water, which correlates to the corresponding CMC
value (Fig. 6).
In vitro digestibility of synthetic sugar alcohol behenates
One of the main objectives for the synthesis of ultra-long fatty
acid sugar alcohol monoesters in this work is to produce exci-
pients for the preparation of solid lipid nanoparticles. Examin-
ation of their ability to undergo lipolysis and their resistance
to digestion is a prerequisite in order to identify them as exci-
pients for SLN preparation for in vivo applications. We thus
designed PPL-catalyzed hydrolysis experiments as an in vitro
lipid digestion model to test the enzymolysis stability of the
compounds, and glycerol palmitic monoester was used as the
control compound. As shown in Fig. 7, all the synthetic com-
pounds could be digested by PPL-catalyzed cleavage of the
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 5 The peak positions of the CH2 and CvO stretches as a function of temperature of compound 1 (a’, c’, d’) and compound 5 (b’, d’, f’).
ester bond to liberate behenic acid. All the synthetic com- limits its water solubility and resistance to hydrolysis. There is
pounds displayed higher resistance to lipolysis compared to no significant difference in enzymatic lipolysis rate between
the control. Compound 2, glycerol behenate, is the natural compounds 3a, 4a, 4b and 5, of which all have enhanced lipo-
substrate of lipase, displaying the highest digestion rate.35 lysis resistance compared to glycerol monobehenate as they
Compounds 1 and 3b showed the lowest digestion rates. The are not the natural substrates of PPL. This lipolysis resistance
reduced ability to undergo lipolysis of ethylene glycol monobe- property of sugar alcohol behenates is an advantage for SLN
henate might be related to its low aqueous solubility, and the excipients, as it could prolong the half-life of SLNs as drug car-
low enzyme digestion rate for compound 3b might be ascribed riers, for controlled release in the body. However, it is worth
to the strong intramolecular bonds in its polar head, which noting that their digestibilities may behave differently when
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
SEM images of the compound 1 and compound 5 SLNs
(shown in Fig. 8c and d, respectively) are in agreement with
the DLS results. In Fig. 8c the arrows point to the three major
types of nanoparticles observed in compound 1-based solid
nanodispersions, which indicates that the nanoparticles have
a very large size distribution, and some larger particles have
very rough surfaces. Meanwhile, compound 5-based SLNs are
rather homogeneous, exhibiting spherical structures with
smooth surfaces. It should be pointed out that these prep-
arations were carried out with no addition of surfactant, which
means compound 5, sorbitol monobehenate, could be an exci-
pient capable forming surfactant-free SLNs. It is thus antici-
pated that this solid lipid could be an excipient that may find
potential applications in the design of new delivery systems.
Fig. 6 CMC values of synthetic compounds and their correlation with
the corresponding melting points. b–h correspond to compounds 1, 2,
3a, 3b, 4a, 4b and 5, respectively.
Prospective multipurpose applications
As an important type of biomaterials, solid lipids have played
an enormous role in the success of cosmetic and food formu-
lation development, as well as in medical devices and drug
delivery systems.3,6,7,9,10,13 Meanwhile the design, synthesis
and development of solid lipids with defined structures for
specific applications always represents a challenge for che-
mists and biochemists.8–13 As seen from the present study,
sugar alcohol monobehenates represent an interesting group
of solid lipids with unique molecular structures: an ultra-long
acyl chain moiety and a polar head with a variable number of
OH groups. The ultra-long alkanoyl chains enable strong
hydrophobic interactions directing chain packing. The
number of OH groups in the polar head and the size and
volume of the sugar alcohol moiety determine the changes in
melting points and surface activity, and perturb the alkanoyl
packing. The structure variations thus endow these molecules
with different chemical and physical properties, and conse-
quently they could be used for different applications. Ethylene
glycol and glycerol monobehenates have a small polar head
and an ultra-long acyl chain. This feature of their molecular
structures facilitates orthorhombic packing, which means they
could find applications in cosmetic product formulation. Mul-
tiple hydroxyl groups in the sorbitol part in sorbitol monobe-
henate give this molecule improved emulsification properties,
and it is found to be capable of forming stable homogeneous
nanoemulsions without additional surfactant. Sorbitol mono-
behenate can be enzymatically lipolysed, but improved resist-
ance to enzymatic lipolysis was observed. This indicates that
sorbitol monobehenate could be an excellent excipient for the
preparation of SLNs. Thus, this work not only adds to the
Fig. 7 The time course of PPL hydrolysis of sugar alcohol monobehe-
nates, with glycerol monopalmitate as a control.
mixed with other excipients in real SLNs with drug molecules
loaded. The real digestive profiles of sugar alcohol behenates
in specific SLN systems can only be mapped as their in vitro
and in vivo digestion in solid lipid nanoparticles are measured,
which would be an interesting topic for further study.
Sugar alcohol monobehenate-based nanoparticles
The synthetic compounds were employed for the preparation development of green synthesis technology, but also expands
of solid lipid nanoparticles by an emulsification–diffusion the molecular library for potential applications in the cos-
technique without additional surfactants. To demonstrate the metic, food and pharmaceutical industries. Furthermore, DCM
concept, two extreme cases, compounds 1 and 5, with the and chloroform, etc., “dirty solvents” used in the lab scale
largest difference in CMC value, were chosen. The DLS analysis preparation in this work, can be replaced by hexane etc. in
shows that the average size of compound 5-based SLNs is scale-up preparation, which may lead to a greener process.
95.41
15.61 nm (Fig. 8a) with a narrow PDI 0.33
0.04.
A more detailed characterization and property evaluation of
However, the size range of the compound 1-based solid nano- the synthetic sugar alcohol monobehenates and further
dispersions was so broad (PDI = 1, data not shown), that the exploiting of a variety of multipurpose applications are in pro-
instrument indicates the sample is out of DLS detection. The gress in our lab.
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 8 Comparisons of SLNs made from compound 1 and compound 5. Particle size distribution of compound 5-based SLNs (a), SLN dispersion
solutions (b), SEM images of SLNs of compound 1 (c) and compound 5 (d).
6 C. Fong, T. Le and C. J. Drummond, Chem. Soc. Rev., 2012,
41, 1297–1322.
Acknowledgements
Wei Wei thanks the Chinese Scholarship Council (CSC) for the
financial support for her visiting stay at Aarhus University.
Technical support from the Department of Engineering and
Interdisciplinary NanoScience Center (iNANO), Aarhus Univer-
sity is gratefully acknowledged.
7 G. John and P. K. Vemula, Soft Matter, 2006, 2, 909–914.
8 A. A. Koutinas, A. Vlysidis, D. Pleissner, N. Kopsahelis,
I. Lopez Garcia, I. K. Kookos, S. Papanikolaou, T. H. Kwan
and C. S. K. Lin, Chem. Soc. Rev., 2014, 43, 2587–2627.
9 P. Foley, A. K. Pour, E. S. Beach and J. B. Zimmerman,
Chem. Soc. Rev., 2012, 41, 1499–1518.
10 R. Langer and D. A. Tirrell, Nature, 2004, 428, 487–492.
11 R. DiCosimo, J. McAuliffe, A. J. Poulose and G. Bohlmann,
Chem. Soc. Rev., 2013, 42, 6437–6474.
Notes and references
12 M. C. R. Franssen, P. Steunenberg, E. L. Scott, H. Zuilhof
and J. P. M. Sanders, Chem. Soc. Rev., 2013, 42, 6491–6533.
13 M. B. Ansorge-Schumacher and O. Thum, Chem. Soc. Rev.,
2013, 42, 6475–6490.
14 M. Miyazawa, T. Takahashi, I. Horibe and R. Ishikawa,
Tetrahedron, 2012, 68, 2007–2010.
15 J. Chopineau, F. D. McCafferty, M. Therisod and
A. M. Klibanov, Biotechnol. Bioeng., 1988, 31, 208–214.
16 J. Piao and S. Adachi, Innovative Food Sci. Emerging
Technol., 2006, 7, 211–216.
17 A. Ducret, A. Giroux, M. Trani and R. Lortie, Biotechnol.
Bioeng., 1995, 48, 214–221.
1 M. Poliakoff and P. Licence, Nature, 2007, 450, 810–812.
2 R. J. White, R. Luque, V. L. Budarin, J. H. Clark and
D. J. Macquarrie, Chem. Soc. Rev., 2009, 38, 481–494.
3 P. T. Anastas and M. M. Kirchhoff, Acc. Chem. Res., 2002,
35, 686–694.
4 M. Poliakoff, J. M. Fitzpatrick, T. R. Farren and
P. T. Anastas, Science, 2002, 297, 807–810.
5 M. I. Hoffert, K. Caldeira, G. Benford, D. R. Criswell,
C. Green, H. Herzog, A. K. Jain, H. S. Kheshgi,
K. S. Lackner, J. S. Lewis, H. D. Lightfoot, W. Manheimer,
J. C. Mankins, M. E. Mauel, L. J. Perkins, M. E. Schlesinger,
T. Volk and T. M. L. Wigley, Science, 2002, 298, 981–987.
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
18 A. Sorrenti, O. Illa and R. M. Ortuno, Chem. Soc. Rev., 2013, 37 J. B. Brubach, V. Jannin, B. Mahler, C. Bourgaux,
42, 8200–8219.
P. Lessieur, P. Roy and M. Ollivon, Int. J. Pharm., 2007, 336,
248–256.
38 E. B. Souto, W. Mehnert and R. H. Müller, J. Micro-
encapsulation, 2006, 23, 417–433.
39 A. Ducret, A. Giroux, M. Trani and R. Lortie, Biotechnol.
Bioeng., 1995, 48, 214–221.
19 J. F. Rabolt, F. C. Burns, N. E. Schlotter and J. D. Swalen,
J. Chem. Phys., 1983, 78, 946–952.
20 C. R. Flach, A. Gericke and R. Mendelsohn, J. Phys. Chem.
B, 1997, 101, 58–65.
21 M. A. Lampe, A. L. Burlingame, J. Whitney, M. L. Williams,
B. E. Brown, E. Roitman and P. M. Elias, J. Lipid Res., 1983, 40 F. Ganske and U. T. Bornscheuer, J. Mol. Catal. B: Enzym.,
24, 120–130. 2005, 36, 40–42.
22 K. P. Ananthapadmanabhan, S. Mukherjee and P. Chandar, 41 H. J. Kim, S. H. Youn and C. S. Shin, J. Biotechnol., 2006,
Int. J. Cosmet. Sci., 2013, 35, 337–345. 123, 174–184.
23 J. A. Bouwstra, P. L. Honeywell-Nguyen, G. S. Gooris and 42 L. Cao, A. Fischer, U. T. Bornscheuer and R. D. Schmid,
M. Ponec, Prog. Lipid Res., 2003, 42, 1–36. Biocatal. Biotransform., 1996, 14, 269–283.
24 J. A. Bouwstra and M. Ponec, Biochim. Biophys. Acta, Bio- 43 E. Castillo, F. Pezzotti, A. Navarro and A. López-Munguía,
membr., 2006, 1758, 2080–2095. J. Biotechnol., 2003, 102, 251–259.
25 A. V. Rawlings and P. J. Matts, J. Invest. Dermatol., 2005, 44 F. Cauglia and P. Canepa, Bioresour. Technol., 2008, 99,
124, 1099–1110. 4065–4072.
26 R. H. Müller, K. Mäder and S. Gohla, Eur. J. Pharm. Bio- 45 U. Hanefeld, L. Gardossi and E. Magner, Chem. Soc. Rev.,
pharm., 2000, 50, 161–177. 2009, 38, 453–468.
27 R. H. Müller, M. Radtke and S. A. Wissing, Adv. Drug Deliv- 46 E. A. Manoel, K. C. Pais, A. G. Cunha, A. B. C. Simas,
ery Rev., 2002, 54(Supplement), S131–S155.
28 W. Mehnert and K. Mäder, Adv. Drug Delivery Rev., 2001,
47, 165–196.
M. A. Z. Coelho and D. M. G. Freire, Org. Process Res. Dev.,
2012, 16, 1378–1384.
47 S. F. Marrian, Chem. Rev., 1948, 43, 149–202.
29 V. Jenning, M. Schäfer-Korting and S. Gohla, J. Controlled 48 C. Demetzos, J. Liposome Res., 2008, 18, 159–173.
Release, 2000, 66, 115–126.
49 A. Badia, L. Cuccia, L. Demers, F. Morin and R. B. Lennox,
J. Am. Chem. Soc., 1997, 119, 2682–2692.
50 W. Mehnert and K. Mäder, Adv. Drug Delivery Rev., 2012,
64(Supplement), 83–101.
30 N. Haagsma, C. M. Gent, J. B. Luten, R. W. Jong and
E. Doorn, J. Am. Oil Chem. Soc., 1982, 59, 117–118.
31 P. A. o. O. A. C. Cunniff, Official methods of analysis of AOAC
international, Association of Official Analytical Chemists, 51 D. J. McClements, Prog. Lipid Res., 2013, 52, 409–423.
Washington, DC, 1995.
32 H. Zhang, G. Önal, C. Wijesundera and X. Xu, Process
Biochem., 2009, 44, 534–539.
52 R. A. H. Lewis and R. McElhaney, in Methods in Membrane
Lipids, ed. A. Dopico, Humana Press, 2007, vol. 400, ch. 14,
pp. 207–226.
33 E. D. Goddard, N. J. Turro, P. L. Kuo and 53 J. Caussin, G. S. Gooris and J. A. Bouwstra, Biochim.
K. P. Ananthapadmanabhan, Langmuir, 1985, 1, 352–355. Biophys. Acta, Biomembr., 2008, 1778, 1517–1524.
34 L. Sek, C. J. H. Porter, A. M. Kaukonen and W. N. Charman, 54 D. J. Moore and M. E. Rerek, Acta Derm.-Venereol., 2000,
J. Pharm. Pharmacol., 2002, 54, 29–41. 16–22.
35 L. Sek, C. J. H. Porter and W. N. Charman, J. Pharm. 55 R. N. A. H. Lewis and R. N. McElhaney, Biochim. Biophys.
Biomed. Anal., 2001, 25, 651–661. Acta, Biomembr., 2013, 1828, 2347–2358.
36 Z. Urbán-Morlán, A. Ganem-Rondero, L. M. Melgoza-Con- 56 D. Chandler, Nature, 2005, 437, 640–647.
treras, J. J. Escobar-Chávez, M. G. Nava-Arzaluz and 57 A. Priev, S. Zalipsky, R. Cohen and Y. Barenholz, Langmuir,
D. Quintanar-Guerrero, Int. J. Nanomed., 2010, 5, 611–620.
2002, 18, 612–617.
This journal is © The Royal Society of Chemistry 2015
Green Chem.