Synthesis and preliminary investigations into norbornane-based amphiphiles and their self-assembly

Article abstract of DOI:10.1039/c3nj00145h

A range of norbornane based amphiphiles, which possess a rigid 'kink' in the centre of amphiphiles, were accessed via a concise four step synthesis. The self-assembly properties of these novel compounds were then investigated and the critical aggregation

Full text of DOI:10.1039/c3nj00145h

NJC
View Article Online
View Journal | View Issue
PAPER
Synthesis and preliminary investigations into
norbornane-based amphiphiles and their
self-assembly†
Cite this: NewJ.Chem., 2013,
37, 1895
a
b
cd
Jennifer S. Squire, Alessandra Sutti, Gregory Durand,* Xavier A. Conlana and
´
Luke C. Henderson*ab
A range of norbornane based amphiphiles, which possess a rigid ‘kink’ in the centre of amphiphiles,
were accessed via a concise four step synthesis. The self-assembly properties of these novel compounds
were then investigated and the critical aggregation concentration (CAC), hydrodynamic diameter (DH)
by dynamic light scattering (DLS) and their morphology by cryogenic transmission electron microscopy
(cryoTEM) and negatively stained transmission electron microscopy (TEM) were determined. These
compounds while possessing similar CAC values (50–70 mM) exhibited a wide variety of particle size
(60–140 nm) and morphologies, including vesicles, cigar-shaped aggregates and rod-like micelles.
Considering the similarities in molecular structure we have proposed that the unique nature of the
molecular ‘kink’ is affecting molecular assembly in which subtle changes in molecular structure have
large ramifications on aggregate size and morphology.
Received (in Victoria, Australia)
5th February 2013,
Accepted 20th March 2013
DOI: 10.1039/c3nj00145h
www.rsc.org/njc
importance as this provides insights into molecular features
which contribute to potentially desirable properties for a given
Introduction
Self-assembly of amphiphilic compounds has been of incred-
ible interest to the scientific community and is at the core of
many cutting-edge technologies in areas of medicine, nutra-
ceuticals, gene-delivery and in the various aspects of materials
science.1–4 It has been noted that even small changes in
molecular structure of the individual amphiphile can have
major ramifications on the morphology of the self-assembled
nanostructures that they produce. Therefore, in the continual
search for more advanced and ‘tailorable’ nanostructures, the
correlation between the molecular structure of amphiphiles
and the characteristics of the aggregates they form is of utmost
system.
Recently, Matisons and co-workers5 stated that the inclusion
of a rigid portion in the centre of an amphiphile could lead to
the formation of lipid bi-layers and, depending on angular
displacement between hydrophobic tails caused by this rigidity,
may lead to the formation of ribbons or tube-like structures.
The rigid groups referred to by Matisons were typically a
conjugated series of unsaturated carbon–carbon bonds placed
between the polar head group and an unconstrained hydro-
phobic tail.
In this work we have installed a norbornane unit within the
core of a range of amphiphiles as this scaffold provides rigidity,
angular displacement while being entirely based on a saturated
hydrocarbon scaffold. We hypothesised that the inclusion of a
norbornane unit at the core of the amphiphile structure should
promote the formation of vesicles due to the aforementioned
effects. The norbornane scaffold, bearing a primary amine, has
been employed as a counter-ion in a study by Bordes at al.6,7 In
this instance the amino portion of the norbornane was a
60/40 mixture of endo–exo and this system demonstrated large
variations in assembly size with minor changes in molecular
feature. This was largely attributed to the unique bicyclic
structure of the norbornane scaffold and its effect on ion
pairing and assembly. The inclusion of the norbornane within
a Strategic Research Centre for Chemistry and Biotechnology, Deakin University,
Geelong, Victoria 3216, Australia. E-mail: luke.henderson@deakin.edu.au;
Fax: +61 3 52271045; Tel: +61 3 52272767
b Institute for Frontier Materials, Deakin University, Geelong, Victoria 3216,
Australia
c
´
Universite d’Avignon et des Pays de Vaucluse, Equipe Chimie Bioorganique et
`
Systemes Amphiphiles, 33 rue Louis Pasteur, F-84000 Avignon, France
d
´
Unite Mixte de Recherche 5247, Centre National de la Recherche Scientifique and
´
´
´
Universites de Montpellier 1&2, Institut des Biomolecules Max Mousseron, Faculte
de Pharmacie, 15 avenue Charles Flahault, F-34093 Montpellier Cedex 05, France
† Electronic supplementary information (ESI) available: 1H and 13C NMR for all
novel compounds, representative CAC and DLS outputs and lipophilicity to
methylene number correlation graphs are presented. See DOI: 10.1039/
c3nj00145h
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 1895--1905 1895

View Article Online
Paper
NJC
the amphiphiles has provided the architecture with what we have
termed a molecular ‘kink’ and will be another area of novelty
explored in this work. Additionally, the synthetic methodology
employed within our group using the norbornane scaffold is
well established.8
In this manuscript we present a highly versatile synthesis for
accessing norbornane based amphiphilic compounds using
reagents which are ubiquitous within organic chemistry
laboratories worldwide. Additionally, we present preliminary
findings on the physicochemical properties of several amphi-
philes which include critical aggregation concentration (CAC),
hydrodynamic diameter (DH) and morphology using dynamic
light scattering (DLS) and cryogenic transmission electronic
microscopy (cryoTEM) and TEM.
Scheme 2 OsO4–NMO mediated dihydroxylation of imides 7–10.
step synthetically viable yields remained elusive; as such this
imide was no longer pursued for further study.
Again, reasonable yields were realised with most compounds,
though it should be noted that the purification of these compounds
proved troublesome.
Results and discussion
Installation of the hydrophobic portion of the amphiphiles
was carried out using a cyclodehydration protocol to give a
1,3-dioxolane (acetal) moiety under mildly acidic conditions.10
These acetals were purified and immediately deprotected by
catalytic hydrogenolysis to give the neutral amphiphiles which
were obtained in good yields over two steps (Scheme 3). The
overall yield for the synthesis of these novel norbornane-based
amphiphiles ranged from 42% to 12% in four steps.
As we were interested in the properties of the amphiphiles
in the cationic state, each of the amphiphiles 11–18 were
converted to their corresponding hydrochloride salt using
gaseous HCl.11 It is worth noting that the described synthetic
route can easily incorporate any amine or aldehyde combi-
nation, whether they are commercially available or tailored for
a specific means. For example, the nitrogen unit used for imide
formation could form part of small peptides, proteins or
aptamers for targeted therapeutic purposes.
Given the unusual structural nature of the amphiphiles
synthesised in this study we were curious about the lipophilic
nature of these compounds. Specifically, we were interested
whether the norbornane core is considered part of the lipo-
philic region and how much the cationic spacer unit versus the
alkyl chain (i.e. n versus m, Fig. 1) contributes to the overall
lipophilicity of these compounds.
When considering the synthetic pathway to give amphiphiles of
general structure shown in Scheme 1, we were conscious of
developing a route that could be easily modified to facilitate
analogue synthesis. We began by using mono-protected alkyl-
diamines,8,9 with which we have had previous experience, of
various lengths to access imides 2–5 using microwave irradia-
tion and a slight excess (1.5 equivalents) of the mono-protected
amine.
This protocol worked well for most mono-protected amines
giving high to excellent yields of the corresponding imides (2, 3
and 5) however, compounds 4 and 6 were initially isolated in
very poor yields (typically o20%). Given these low yields brief
microwave optimisations were investigated for these two
amines, slightly increasing the reaction time, temperature
and equivalents of amine (55 min, 120 1C, 2 eq. of amine,
highlighted in Scheme 1 with asterisk) these compounds could
be isolated in synthetically useful yields shown in Scheme 1
(for optimisation table refer to ESI†).
With imides 2–5 in hand, our attention turned to installing
the hydrophobic portion of the amphiphiles. Imides 2–5 were
dihydroxylated using a typical osmium tetroxide–N-methyl-
morpholine-N-oxide (OsO4–NMO) in acetone system to exclu-
sively give the exo-vicinal diols 7–10 (Scheme 2),8 the selective
formation of the diol moiety in this orientation is imperative to
the overarching ‘kink’ in the amphiphile. Note that treatment
of 6 under these dihydroxylation conditions gave very low yields
(typically 10–20%). Despite attempts at optimisation of this
Scheme 3 Installation of linear hydrophobic chain via an acetal moiety and
Scheme 1 Synthesis of imides 2–6.
removal of the Cbz protecting group.
c
1896 New J. Chem., 2013, 37, 1895--1905
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Paper
Fig. 1 Core scaffold for evaluation highlighting m and n spacer units.
Table 1 Predicted and determined lipophilicity values
b
nCa
log P
log kw
0
Entry
Compound
n
m
1
2
3
4
5
6
7
8
Norbornane
11
12
13
14
15
16
17
—
1
1
2
2
4
4
5
5
—
—
—
—
—
10
14
10
14
10
14
10
14
—
—
—
—
—
13
17
15
19
19
23
21
25
12
14
16
18
ꢀ1.65c
1.77
3.28
2.00
3.44
3.13
4.22
3.66
4.65
1.64
2.53
3.26
3.95
—
1.01
1.13
0.89
1.18
1.25
1.30
1.33
1.33
—
Fig. 2 Graphical representation of A log P values versus number of methylene
units.
9
18
compound 11 (Fig. 2). A similar scenario was observed when
comparing the n value in compounds 12, 14, 16 and 18 (Fig. 2).
10
11
12
13
C12H23NH3Cl
C14H29NH3Cl
C16H33NH3Cl
C18H37NH3Cl
—
—
—
0
Note that a correlation of both the log P and log kw with the
number of methylene units present in these compounds has
Number of alkyl carbons within amphiphiles, for norbornane-based been undertaken and is presented in the ESI.†
a
compounds nC = (2n + m + 1), the acetal carbon being omitted.
These differences suggest that the choice of diamine which
is used for the synthesis has a much less pivotal role in
determining overall lipophilicy of the target compound. The
b
c
Determined by RP-HPLC. Data from ref. 7.
Therefore, the partition coefficients of 11–18 were predicted general trends observed by comparison of log P are supported
using the software ALOGPS12,13 and the values are given in by log kw though, again, the magnitude of variation is much
Table 1. The norbornane (Table 1, entry 1) is reported here as smaller in the latter. Disparate results between the classical
an indication of what the core unit lipophilic properties are and octanol partitioning method and the HPLC methods have been
thus how the inclusion of separation between the units m and n observed for short cationic peptides, most likely because of
have on this overall characteristic. In addition to the calculated interactions with the stationary phase of the column. Due to the
values we experimentally determined the log kw0, a parameter cationic nature of the polar heads, similar interactions may
closely associated with log P which is commonly used for small explain the variation between the predicted and the experi-
molecules and amphiphiles.14 For the sake of comparison, with mental values.15
0
respect to log P linear n-alkylammonium chloride compounds
Comparing the lipophilicity of the norbornane-based amphi-
were added as well (Table 1, entries 10–13).
philes to that of their linear n-alkyl ammonium chloride deriva-
In comparing the partition coefficient for compounds 11 tives of similar chain length, it appears that the norbornane
and 12, it can be seen that extension of the alkyl chain installed scaffold does not contribute much to the overall lipophilicity
via the acetal moiety (m) by four methylene units results in a despite being mostly hydrocarbon based (Table 1). A reason for
significant increase of the overall lipophilicity by B1.9 times. the lack of lipophilicity demonstrated by the norbornane core
A similar scenario is observed when comparing compounds 13 may be due to the presence of the polar oxygens within the acetal
and 14, compounds 15 and 16 and compounds 17 and 18, and carbonyls at each end of the scaffold; this in effect may
where installation of four carbons on the aliphatic chain has negate the lipophilic hydrocarbon core of the bicycle.
corresponded to an increase of the lipophilicity by B1.7, 1.3
Indeed n-dodecylammonium chloride exhibits a log P value
and 1.3 times, respectively. This clearly shows that the pre- (1.64) very close to that of compound 11 (1.77) having 13 alkyl
dicted effect of extending the alkyl chain on the lipophilicity is carbons. With longer alkyl chains, more pronounced differ-
more pronounced when the diamino spacer is short (Fig. 2).
ences are observed as the n-octadecylammonium chloride
Comparison of these same compounds (11 and 12, 13 and exhibits a higher lipophilicity than compounds 14 and 15
0
14, 15 and 16, 17 and 18) with respect to the determined log kw
although both have an overall number of alkyl carbons of 19
values shows the same trend though the variation in lipo- suggesting that the norbornane scaffold is not lipophilic by
philicity is much smaller. itself. With this data in hand our attention turned to an
Continuing with this theme, analysis of compounds, 11 and investigation of aggregation properties focusing on compounds
13, increasing n by two carbons shows only a mild increase in 11–14. The selection of these four compounds provides a
log P value (B1.1 times) whereas, as expected, six (15) or eight comparison of alkyl chain length and amine spacer unit when
(17) carbons of the spacer length led to much more lipophilic incorporated into this novel scaffold, additionally compounds
compounds, respectively, 1.8 and 2.1 times when compared to 15–18 proved very insoluble in aqueous solution.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 1895--1905 1897

View Article Online
Paper
NJC
Determination of critical aggregation concentration (CAC) by
encapsulation of pyrene
were found to form aggregates at a threshold concentration
of B12 mM using fluorescence technique.25 Therefore, this
suggests that compounds 11–14 having CAC values ranging from
50 to 70 mM behave more or less like a C16 surfactant with no
significant effect of n and m on the aggregation concentration.
When considering the effect of the length of the diamino
spacer, n, a singular behaviour was also observed as for a given
alkyl chain, similar CAC values were observed for the butyl
diamino derivatives (13 and 14) when compared to the ethyl
diamino ones (11 and 12). This is, however, in agreement with
the partition coefficients as we showed that increasing the
length of the spacer unit from 2C to 4C did not significantly
affect the overall lipophilicity (Table 1). This, again, correlates
well to our previous hypothesis that the length of the diamino
spacer incorporated into this scaffold does not play a pivotal
role in overall compound hydrophobicity.
The critical aggregation concentration (CAC) of amphiphilic
species is of paramount importance when considering
their application to biological systems. In our hands, these
compounds proved troublesome to dissolve in water but
adequate solubilisation was achieved by heating at 60 1C under
ultrasonic agitation for an hour (see experimental section for
details). Once dissolved in water the CAC was determined by
monitoring the ratio of fluorescence emitted by the encapsula-
tion of pyrene.16,17
The ratio of pyrene emission peaks at 386 nm and 393 nm
(I386/393) was monitored over a series of concentrations ranging
from 1–0.0001 mg mLꢀ1, as per literature procedures.16,18 An
example of the CAC determination of compound 11 is shown in
the ESI† (Page S18), where a sharp increase in the I386/393 value
can be seen at 30.8 mg Lꢀ1 (70 mM) suggesting that this is the
concentration at which aggregation has occurred, allowing the
encapsulation of pyrene in the aggregate core. The CAC data for
all compounds are summarised in Table 2.
This further confirms the peculiar behaviour of these com-
pounds as they exhibit similar critical aggregation concentra-
tions to that of C16 surfactants.24,25 Moreover, such a magnitude
of concentration is consistent with other single-chain surfactants
forming bilayer aggregates.19,25
As shown in Table 2, determination of CAC demonstrated
We have attributed these unusual results, which contradict
commonly observed trends, to the unique architecture which is
present in these amphiphiles. Indeed, due to the weakly non-
polar nature of the norbornane core as discussed previously
(see partition coefficient section), as well as its rigid structure
and the kink it provides to the overall molecular architecture,
its insertion within the hydrophobic part of the amphiphile
may alter aggregation and therefore could explain unexpected
behaviour. For instance, although the sulfur atom is usually
considered as a hydrophobic unit when inserted within the
hydrophobic chain of a surfactant, Menger and co-workers
found out that the insertion of a sulfide group causes an
increase in the CAC by 2- to 3-fold, the deeper the insertion,
the higher the increase.26,27
Taking this into account, on one hand, an increased CAC
may be explained by attractive interactions between the
norbornane core and water such as hydrogen bonding leading
to increased solubility. The norbornane group could also
decrease the entropic gain on micellisation by changing the
water structure around the monomer into a less ordered one.
On the other hand, packing constraints could arise from the
rigid and bulky norbornane inserted group. These effects will
be further studied in future work.
0
behaviour consistent with log kw data as the CAC values
exhibited by compounds 11–14 were fairly insensitive to
changes made to amphiphile structure. This observation corre-
lates well with those reported by Kunitake et al.19 who also used
a rigid scaffold in simple ammonium-based single chain
amphiphiles.16 When comparing 11 and 12, only a slight
decrease of the CAC, 22.6 mg Lꢀ1 (50 mM), was observed for
the latter although it is has an additional four methylene longer
alkyl chain.
As frequently reported for a given polar head group, the
addition of two methylene units to the hydrocarbon chain of a
surfactant decreases 5–15 times the critical aggregation concen-
tration.20,21 A similar observation was made with compounds
13 and 14, the more hydrophobic derivative exhibited a CAC
value only 1.2 times lower than that of 13 (70 mM vs. 60 mM,
respectively) which possesses four fewer carbons. When com-
paring the CAC of compounds 11–14 to that of linear n-alkyl
ammonium compounds, we found out that they were somehow
in relative good agreement with a C16 surfactant. Indeed, the
CAC of n-decyl ammonium chloride was reported to be B0.05
M22 while that of its n-tetradecyl derivative is B3 mM23 and
that of the hexadecyl ammonium bromide B30 mM.24 Similarly,
n-hexadecyl-D-maltosylamine and n-hexadecyl-D-lactosylamine
Determination of hydrodynamic diameter (DH) by DLS
The hydrodynamic diameter of each amphiphile solution was
determined by Dynamic Light Scattering (DLS) at 25 1C. Hydro-
dynamic diameter for 11 showed a DH of 59 nm. Examination of
12 in an aqueous system showed a bimodal distribution
(Table 3, entry 2) of 79 and 24 nm, these data suggest that
the aggregates formed by 12 are in two distinct populations.
DLS analysis of 13 and 14, gave polydisperse aggregates of 142
and 91 nm, respectively.
Table 2 CAC data for 11–14 via pyrene encapsulation
CACa
(mg Lꢀ1
)
(mmol Lꢀ1
)
Entry
Cpds
n
m
nC
1
2
3
4
11
12
13
14
1
1
2
2
10
14
10
14
13
17
15
19
30.8 2.1
22.6 3.1
34.4 1.1
29.3 3.7
70 5
50 7
70 2
60 7
a
Aggregates of B60–160 nm diameter is consistent with
the formation of vesicles as it has been observed with other
Results are the average of triplicate experiments carried out at 18 1C in
MilliQ water.
c
1898 New J. Chem., 2013, 37, 1895--1905
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Paper
Table 3 DH of aggregates
a
Entry
Compound
DH (nm)
1
2
11
12
59 (100%)
79 (89%)
24 (10%)
3
4
13
14
142 (100%)
91 (100%)
a
Hydrodynamic diameter, determined at 10ꢁ the CAC value at 25 1C
with volume distribution.
Fig. 4 Compound 12 showing two morphologies, A, cigar shaped vesicle
(57 nm wide, 163 nm long); B, rod-like micelle (20 nm wide, 245 nm long).
single-chain surfactants.19,25,28 For instance,
a
recently
designed triazole-based single-chain surfactant was found to
form vesicles whose size varies from 80–100 nm by TEM to
B170 nm by DLS.29 Moreover, Bhattacharya and Acharya
reported that freshly prepared solution of the single chain
n-hexadecyl-D-maltosylamine and n-hexadecyl-D-lactosylamine
compounds form vesicles of 50 to 80 nm diameter. Upon
aging, these vesicles transformed into tubular and lamellar
microstructures.25
Considering the data obtained by DLS, our attention turned
to obtaining morphological information of these aggregates
under cryoTEM conditions.
Fig. 5 CryoTEM images of 13 showing the formation of vesicles ranging from
150–250 nm.
TEM and CryoTEM imaging
Compound 11 presented in the form of small vesicle aggregates
(Fig. 3) which were visualised by negative stain. Despite
potential morphological changes that can occur by negative
staining, there was excellent agreement between the DH obtained
via DLS and the observed vesicles, cf. 59 nm vs. 65 nm,
respectively.
Analysis of 12 (Fig. 4) gave a bimodal distribution in DLS
and proved very interesting when visualized by cryoTEM
showing two morphological forms, both of which were cylind-
rical. Examination in low contrast revealed elongated forms of
cigar-shaped vesicles as seen in Fig. 4 (labelled A). These
elongated vesicles were the dominant feature of the colloid,
though a number of striated tube-like structures were also seen,
although these were not as common as the cigar like vesicles
but are still considered to be a real feature of the sample (Fig. 4,
labelled B).
Fig. 6 Compound 14 showing vesicle formation ranging from 70–90 nm.
of their norbornane bicyclic structure. Compound 14 also
showed classic spherical vesicles (Fig. 6) which are quite poly-
disperse in size, with a range of 30–500 nm, however the
majority are observed to be between 70–90 nm, corresponding
well to the values obtained from DLS (cf. 91 nm). Most of the
vesicles are smooth with no substructure although occasional
smaller vesicles observed with a slight ‘‘orange peel’’ texture.
We have attributed these differences in morphology to the
ability of the amphiphiles to stack together, thus creating the
lipid bilayer. It has been shown that the position of the ‘‘rigid
portion’’ of amphiphiles can have ramifications on the overall
morphology of the self assemblies.30 In this study by varying
the value of n and m we have effectively moved the rigid ‘kink’
within the overall architecture of the amphiphiles. As such their
ability to stack must be determined by a delicate balance of n
and m which, in turn, dictates the aggregate morphology, CAC
and size. It was noted that during the course of the CAC studies
Interestingly, 13 has formed large vesicles of classical lipo-
somal structure (Fig. 5), a morphology very similar to that
observed by Bordes et al.7 which was attributed to the stacking
Fig. 3 Compound 11 negatively stained showing vesicles approximately 65 nm
in diameter.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 1895--1905 1899

View Article Online
Paper
NJC
via surface tension measurements that the calculated head using Milli-Q water (resistivity of 18.2 MO cm) at a range of
group area at the interface is larger than typically seen in these concentrations (1–0.0001 mg mLꢀ1). Excess crystalline pyrene
systems (data not shown). This may indicate unusual inter- was then added to the surfactant aqueous solutions and heated
actions at the surface, possibly arising due to stacking phenom- at 30 1C for 30 min in a sonic bath before allowing to
ena arising from the kink present in the amphiphile. The equilibrate at room temperature 23 hours. The solutions were
assembly of these compounds may occur in a complementary then filtered through a 0.45 mM filter prior to measurement.
fashion, with all the kinks facing in the same direction or, The emission spectra of pyrene were acquired by exciting
conversely, in an opposing orientation. Each of these assem- samples at 335 nm (Ex slit width 5 nm, Em slit width 5 nm).
bling motifs will result in vastly different effective critical The spectra were then used to determine the ratio of I386/393.
packing parameter (CPP)31 thus shedding light on the unusual
0
Determination of log kw
behaviour of these compounds. The molecular interactions of
these compounds and their potential assembly modes are The analytes were sequentially injected onto the column
currently under investigation using computational methods with increasing amounts of organic modifier in the eluent
and will be reported in due course.
(water–methanol). Runs were undertaken at ratios of 90 : 10,
70 : 30 and 50 : 50 methanol–water. The value of kw was deter-
mined by extrapolation of the k vs. conc. of organic modifier
back to pure water. The log (base 10) of this value was then
Conclusions
In conclusion, this manuscript has presented the synthesis and carried out to give the log kw value. Dead time of the system was
preliminary investigations into the self-assembly characteris- determined by injecting acetone through the system and mon-
tics of several amphiphiles based on a norbornane scaffold. itoring its retention time.
This scaffold, while rigid in structure, provides a central ‘kink’
to the amphiphile which has a major effect on how they
CryoTEM and TEM
assemble in aqueous media. Prediction of partition coefficient A laboratory-built humidity-controlled vitrification system was
values in silico suggested that the rigid norbornane core has used to prepare the samples for Cryo-TEM. Humidity was kept
little-to-no effect on overall compound hydrophobicity. This close to 80% for all experiments, and ambient temperature was
0
was reinforced by the determination of log kw which showed 22 1C. 200-Mesh copper grids coated with perforated carbon
very small variation in lipophlicity among compounds investi- film (Lacey carbon film: ProSciTech, Qld, Australia) were glow
gated. Additionally the length of the linking group between the discharged in nitrogen to render them hydrophilic. 4 mL
norbornane core and hydrophilic head group has minimal Aliquots of the sample were pipetted onto each grid prior to
effect on CAC value and aggregate morphology as is consistent plunging. After 30 seconds adsorption time the grid was blotted
0
with log kw values. Despite compounds 11–14 possessing manually using Whatman 541 filter paper, for approximately
similar CAC values they demonstrated large variations in 2 seconds. Blotting time was optimised for each sample. The
aggregate size and morphology. Subtle changes in the distribu- grid was then plunged into liquid ethane cooled by liquid nitrogen.
tion of methylene groups throughout the compounds influenced Frozen grids were stored in liquid nitrogen until required. The
the aggregate size (approximately 60–140 nm) and their morpho- samples were examined using a Gatan 626 cryoholder (Gatan,
logy in aqueous solution.
Pleasanton, CA, USA) and Tecnai 12 Transmission Electron
Microscope (FEI, Eindhoven, The Netherlands) at an operating
voltage of 120 KV. At all times low dose procedures were
followed, using an electron dose of 8–10 electrons per Å2 for
all imaging. Images were recorded using an Eagle 4k ꢁ 4k CCD
Experimental
Dynamic light scattering
The light scattering measurements were performed with a camera (FEI, Eindhoven, The Netherlands) using magnifica-
Zetasizer Nano ZS (Malvern Instruments) at a scattering angle tions in the range 15 000–60 000ꢁ
of 1731. The concentrations of the surfactants were sufficiently
Synthesis of N-(benzyloxy carbonyl)-1,4-butanediamine
low enough (0.46–0.72 mM) to avoid multiple scattering from
the aggregates. Solutions were prepared at 10 ꢁ CAC from a A solution of 1,4-diaminobutane (5.72 mL, 5.67 mmol) in
stock solution of 1 mg mLꢀ1 in Milli-Q water and sonicated for CH2Cl2 (100 mL) was cooled to 0 1C and a solution of benzyl-
30 min to ensure adequate dissolving before being left to chloroformate (1.62 mL, 11.3 mmol) and CH2Cl2 (150 mL) was
equilibrate overnight. All solutions were filtered with 0.45 mm added dropwise over 1 h. The reaction was then stirred at rt for
filter directly into the cuvette immediately before measurement 24 h. The resulting mixture was transferred to a separating
to avoid interference from dust particles. Measurements were funnel where it was washed with saturated aqueous NaCl
taken at 25 1C unless otherwise stated.
(3 ꢁ 30 mL). The organic phase was dried (MgSO4), filtered
and solvents removed in vacuo to afford a white powder.
1H NMR spectroscopy data was found to be consistent
Pyrene encapsulation measurements
The fluorescence properties of pyrene were determined using a with literature values8 for the desired monobenzylcarbamate
Varian Cary Eclipse fluorescence spectrophotometer. The sur- diamine (2.07 g, 82%) in >95% purity which was then used
factant solutions were prepared 24 h prior to the measurements without further purification.
c
1900 New J. Chem., 2013, 37, 1895--1905
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Paper
Synthesis of N-(benzyloxy carbonyl)-1,2-diaminoethane
d 7.34–7.19 (5H, s), 6.01 (2H, s), 5.01 (2H, s,), 3.28–3.26 (4H, m),
3.18–3.08 (4H, m), 1.73–1.68 (1H, d, J = 13.5 Hz), 1.52–1.42 (5H, m,
2 ꢁ CH2); 13C NMR (67.5 MHz, CDCl3): d 177.87, 156.59–156.49,
136.71, 134.51, 129.91–128.16, 66.82–66.64, 52.31, 45.79, 44.96,
40.56, 37.91, 27.31, 25.16; HRMS (ESI m/z): calcd for
[C21H25N2O4]+ 369.1809, found 369.1837; n(max) cmꢀ1: 3338 m,
2942 m, 2870 m, 1683 s, 1527 m, 1172 m.
Synthesis of N-(benzyloxy carbonyl)-1,2-diaminoethane was
carried out as per previously described for N-(benzyloxy carbonyl)-
1,4-butanediamine, with 1,2-diaminoethane (6.67 mL, 10 mmol)
and benzylchloroformate (2.85 mL, 20 mmol mol) to afford a
white paste. 1H NMR spectroscopy data was found to be
consistent with literature values8 for the desired monobenzyl-
carbamate diamine (3.48 g, 90%) in >95% purity which was
then used without further purification.
Synthesis of benzyl [2-(1,3-dioxo-1,3,3a,4,7,7a-hexahydro-2H-
4,7-methanoisoindol-2-yl)ethyl]carbamate 2
Synthesis of N-(benzyloxy carbonyl)-1,8-octyldiamine
Synthesis of imide 2 was carried out as previously describe for
compound 3 with mono-protected diaminoethane (2 g, 0.01 mol)
and norbornene anhydride (1.13 g, 7.0 mmol) to produce a
Synthesis of N-(benzyloxy carbonyl)-1,8-octyldiamine was carried
out as per previously described N-(benzyloxy carbonyl)-1,4-
butanediamine, with 1,8-diaminooctane (3.00 g, 0.021 mol)
and benzylchloroformate (0.6 mL, 0.0042 mol) to afford a white
powder. 1H NMR spectroscopy data was found to be consistent
with literature values8 for the desired monobenzylcarbamate
diamine (0.985 g, 84%) in >95% purity which was then used
without further purification.
1
yellow oil. H NMR spectroscopy analysis showed it to be the
desired diaminoethane imide 2 (2.2 g, 94%) in >95% purity
which was then used without further purification. 1H NMR
(270 MHz, CDCl3): d 7.47–7.31 (5H, s) 6.03 (2H, s) 5.04 (2H, s)
3.46–3.48 (2H, d, J = 5.4 Hz) 3.27–3.32 (2H, m) 3.17 (2H, s)
1.67–1.69 (1H, d, J = 5.4 Hz) 1.46–1.49 (1H, d, J = 8.1 Hz);
13C NMR (67.5 MHz, CDCl3): d 177.96, 156.35, 136.63, 134.51,
128.57–128.17, 66.77, 52.31, 45.87, 44.94, 37.80; HRMS (ESI m/z):
Synthesis of N-(benzyloxy carbonyl)-1,10-decyldiamine
calcd for [C19H21N2O4]+ 341.1496, found 341.1478; n(max) cmꢀ1
:
Synthesis of N-(benzyloxy carbonyl)-1,10-decyldiamine was carried
out as per previously described for N-(benzyloxy carbonyl)-1,4-
butanediamine, with 1,10-diaminodecane (1.00 g, 0.0058 mol)
and benzylchloroformate (0.17 mL, 0.0012 mol) to afford a
white powder. 1H NMR spectroscopy data was found to be
3349 m, 2945 m, 1691 s, 1525 m, 1188 m.
Synthesis of benzyl [2-(1,3-dioxo-1,3,3a,4,7,7a-hexahydro-2H-
4,7-methanoisoindol-2-yl)octyl]carbamate 4
consistent with literature values8 for the desired monobenzyl- Synthesis of imide 4 was carried out as previously describe for
carbamate diamine (0.208 g, 57%) in >95% purity which was compound 3 with mono-protected diaminooctane (100 mg,
then used without further purification.
0.359 mmol) and norbornene anhydride (29.5 mg, 0.180 mmol)
to produce a yellow oil. H NMR spectroscopy analysis showed
1
Synthesis of N-(benzyloxy carbonyl)-1,12-dodecyldiamine
it to be the desired diaminooctane imide 4 (49 mg, 64%) in
>95% purity which was then used without further purification.
1H NMR (270 MHz, CDCl3): d 7.33–7.35 (5H, s), 6.15 (2H, s), 5.10
(2H, s), 3.21–3.56 (6H, m), 1.69–1.72 (1H, d, J = 8.1 Hz), 1.43–
1.48 (5H, m), 1.22 (8H, s); 13C NMR (67.5 MHz, CDCl3): d 177.51,
156.20, 136.49, 128.80, 128.37, 65.99, 52.30, 45.80, 44.98, 41.34,
38.34, 29.18, 26.85, 26.68; HRMS (ESI m/z): calcd for
[C25H33N2O4]+ 425.2435 found 425.2428; n(max) cmꢀ1: 3333 m,
2935 m, 2854 m, 1687 s, 1558 m, 1137 m.
Synthesis of N-(benzyloxy carbonyl)-1,12-dodecyldiamine was
carried out as per previously described for N-(benzyloxy carbonyl)-
1,4-butanediamine, with 1,10-diaminodecane (2.00 g, 0.010 mol)
and benzylchloroformate (0.28 mL, 0.002 mol) to afford a white
powder. 1H NMR spectroscopy data was found to be consistent
with literature values8 for the desired monobenzylcarbamate
diamine (0.556 g, 83%) in >95% purity which was then used
without further purification.
Synthesis of benzyl [2-(1,3-dioxo-1,3,3a,4,7,7a-hexahydro-2H-
Synthesis of benzyl [2-(1,3-dioxo-1,3,3a,4,7,7a-hexahydro-2H-
4,7-methanoisoindol-2-yl)butyl]carbamate 3
4,7-methanoisoindol-2-yl)dodecanyl]carbamate 5
To a 35 mL microwave vial containing mono-protected diamino- Synthesis of imide 5 was carried out as previously describe for
butane (873 mg, 3.90 mmol), norbornene anhydride (430 mg, compound 3 with mono-protected diaminodecane (1.05 g,
2.6 mmol) was added and dissolved with toluene (15 mL). The 0.003 mol) and norbornene anhydride (563 mg, 0.002 mol) to
1
solution was then subjected to microwave irradiation at 100 1C produce a yellow oil. H NMR spectroscopy analysis showed it
for 30 min. The resulting solution was diluted with CH2Cl2 to be the desired diaminodecane imide 5 (1.22 g, 79%) in >95%
(30 mL) and was transferred to a separating funnel where it was purity which was then used without further purification.
washed with saturated aqueous NaCl (2 ꢁ 25 mL) followed by a 1H NMR (270 MHz, CDCl3): d 7.32–7.30 (5H, s), 6.11 (2H, s),
wash with HCl (2 M, 20 mL) and finally NaHCO3 (3 ꢁ 25 mL). 5.05 (2H, s), 3.34–3.18 (8H, m), 1.68–1.71 (1H, d, J = 8.1 Hz),
The combined organic phases were dried (MgSO4) and solvent 1.47–1.45 (5H, m), 1.23 (10H, s); 13C NMR (67.5 MHz, CDCl3):
removed in vacuo to give reddish brown viscous oil. 1H NMR d 177.86, 156.48, 137.01, 134.47, 128.57–128.12, 66.59, 52.27,
spectroscopy analysis showed it to be the desired diamino- 45.77, 44.96, 38.49, 29.42, 29.36, 26.92; HRMS (ESI m/z): calcd
butane imide 3 (1.29 g, 90%) in >95% purity which was then for [C27H37N2O4]+ 453.2748, found 453.2763; n(max) cmꢀ1
used without further purification. H NMR (270 MHz, CDCl3): 3348 m, 2926 m, 2854 m, 1692 s, 1531 m, 1129 m.
:
1
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 1895--1905 1901

View Article Online
Paper
NJC
Synthesis of benzyl [2-(5,6-dihydroxy-1,3-dioxooctahydro-2H-
4,7-methanoisoindol-2-yl)butyl]carbamate 8
1.52–1.47 (4H, d, J = 20 Hz), 1.28 (2H, s), 1.25 (8H, s);
13C NMR (67.5 MHz, CDCl3):
d 177.04, 156.85, 136.47,
128.63–128.23, 70.26, 66.93, 46.01, 45.76, 40.77, 36.02, 29.85,
28.60, 27.80, 26.63, 26.01 HRMS (ESI m/z): calcd for
[C25H35N2O6]+ 459.2490, found 459.2475; n(max) cmꢀ1: 3347 br,
2929 m, 2856 m, 1692 s, 1534 m, 1139 m.
Diaminobutane imide 3 (1.03 g, 2.70 mmol) was dissolved in a
4 : 1 solution of acetone–water (30 mL) and NMO (493 mg,
4.0 mmol) was added and stirred until dissolved. Osmium
tetroxide (0.3 mL, 4% in H2O) was added and the black solution
stirred for 72 h at room temperature. The reaction was
quenched with sodium metabisulfite (2 mL, 0.53 M) and the
solution diluted with EtOAc (30 mL) before being transferred to
a separating funnel where it was washed with saturated aqueous
NaCl (3 ꢁ 25 mL). The organic phase was dried (MgSO4) and
solvent was removed in vacuo to afford a dark brown oil.
Purification by silica gel column chromatography was per-
formed with 80% EtOAc–20% petroleum spirit solution and
Synthesis of benzyl [2-(5,6-dihydroxy-1,3-dioxooctahydro-2H-
4,7-methanoisoindol-2-yl)dodecanyl]carbamate 10
Synthesis of diaminodecane diol 10 was carried out as pre-
viously describe for compound 8 with diaminodecane imide 5
(673 mg, 1.5 mmol) and NMO (262 mg, 2.23 mmol) to produce a
1
black oil which was shown by H NMR spectroscopy to be the
desired diaminodecane diol 10 (334 mg, 46%). Numerous
attempts at purification by silica gel column chromatography
were unsuccessful so the crude product was used without
further purification. 1H NMR (270 MHz, CDCl3): 1H NMR
(270 MHz, CDCl3): d 7.32–7.29 (5H, m), 5.08 (2H, s), 3.66 (3H,
br s), 3.40–3.34 (2H, t, J = 8.1 Hz), 2.64–2.62 (2H, m), 2.01–1.97
(2H, br s), 1.44–1.41 (7H, m), 1.22–1.16 (13H, m); 13C NMR
(67.5 MHz, CDCl3): d 177.13, 156.67, 136.64, 128.60–128.19,
70.15, 66.74, 45.97, 38.76, 35.97, 29.93–29.07, 28.91, 26.86
HRMS (ESI m/z): calcd for [C27H38N2NaO6]+ 509.2622, found
509.2614; n(max) cmꢀ1: 3350 br, 2926 m, 2854 m, 1685 s, 1533 m,
1151 m.
1
the resulting brown oil was shown by H NMR spectroscopy to
1
be the desired diaminobutane diol 8 (733 mg, 67%). H NMR
(270 MHz, CDCl3): d 7.30–7.27 (5H, s), 5.01 (2H, s), 3.98 (2H, s),
3.64 (2H, s), 3.44–3.39 (2H, t, J = 8.1 Hz), 3.11–3.09 (2H, d, J = 5.4 Hz),
2.63 (2H, s), 2.39 (2H, br s), 2.08–2.04 (2H, d, J = 10.8 Hz),
1.48–1.40 (4H, m); 13C NMR (67.5 MHz, CDCl3): d 177.30,
156.85, 136.40, 128.64, 70.19, 66.92, 45.82, 45.75, 40.68, 38.27,
36.00, 29.75, 27.55, 25.16 HRMS (ESI m/z): calcd for
[C21H26N2NaO6]+ 425.1683, found 425.1657; n(max) cmꢀ1: 3368 br,
2942 m, 1689 s, 1537 m, 1139 m.
Synthesis of benzyl [2-(5,6-dihydroxy-1,3-dioxooctahydro-2H-
4,7-methanoisoindol-2-yl)ethyl]carbamate 7
Synthesis of benzyl [2-(2-dodecyl-5,7-dioxooctahydro-6H-4,8-
methano[1,3]dioxolo[4,5-f]isoindol-6-yl)butyl]carbamate and
the corresponding deprotected amphiphile 13
Synthesis of diaminoethane diol 7 was carried out as previously
describe for compound 8 with diaminoethane imide 2 (1.04 g,
3.0 mmol) and NMO (539 mg, 4.6 mmol) to produce a dark
brown oil. Purification by silica gel column chromatography
was performed with 90% EtOAc–10% petroleum spirit solution
and the resulting light brown oil was shown by 1H NMR
spectroscopy to be the desired diaminoethane diol 7 (578 mg,
To a solution of dodecyl aldehyde (0.3 mL, 1.36 mmol) in
CH2Cl2 (30 mL), diaminobutane diol 8 (366 mg, 0.909 mmol)
was added and stirred until dissolved. MgSO4 was added,
followed by the addition of p-toluenesulfonic acid (692 mg,
3.64 mmol). The resulting solution was warmed to 35 1C and
stirred for 24 h before being filtered and the solvent was
removed in vacuo to give a light brown oil. Purification by silica
gel column chromatography was performed with a 70% petro-
leum spirit–30% EtOAc solution and the resulting caramel oil
1
51%). H NMR (270 MHz, CDCl3): d 7.32 (5H, s), 5.09 (2H, s),
3.64 (2H, s), 3.58–3.55 (4H, m), 3.37–3.35 (2H, d, J = 5.4 Hz), 2.90
(2H, s), 2.61 (2H, s), 2.11–2.09 (1H, d, J = 5.4 Hz), 1.43–1.39 (1H,
t, J = 5.4 Hz); 13C NMR (67.5 MHz, CDCl3): d 177.54, 156.92,
136.39, 128.63–128.32, 70.15, 67.12, 45.76, 39.22, 35.99 HRMS
(ESI m/z): calcd for [C19H23N2O6]+ 375.1551, found 375.1578;
1
was shown by H NMR spectroscopy analysis showed it to be
the desired diaminobutane dodecyl acetal (314 mg, 61%) in
>95% purity which was then used immediately in the next
n
(max) cmꢀ1: 3368 br, 2924 m, 2854 m, 1692 s, 1533 m, 1145 m.
1
deprotection step. H NMR (270 MHz, CDCl3): d 7.33 (5H, s),
5.07 (2H, s), 4.63–4.60 (1H, t, J = 2.7 Hz), 3.84 (2H, s), 3.43 (2H,
d, J = 5.4 Hz, 2 ꢁ CH), 3.20–3.18 (2H, d, J = 5.4 Hz), 3.07 (2H, s,
2 ꢁ CH), 2.83 (2H, s), 2.35–2.29 (1H, t, J = 16.2 Hz), 2.01–1.98
Synthesis of benzyl [2-(5,6-dihydroxy-1,3-dioxooctahydro-2H-
4,7-methanoisoindol-2-yl)octyl]carbamate 9
Synthesis of diaminooctane diol 9 was carried out as previously (1H, d, J = 8.1 Hz), 1.61–1.51 (6H, m, 3 ꢁ CH2), 1.24 (18H, bs),
describe for compound 8 with diaminooctane imide 4 (228 mg, 0.88 (3H, t, J = 13.5 Hz, CH3); 13C NMR (67.5 MHz, CDCl3):
0.54 mmol) and NMO (94.4 mg, 0.81 mmol) to produce a dark d 176.4, 156.5, 136.6, 128.6–128.20, 104.2, 101.8, 77.5–76.64,
brown oil which was then used without further purification. 66.8, 44.5, 42.7, 40.5, 38.2, 37.4, 35.9, 32.7, 31.9, 29.8–29.2, 27.5,
Purification by silica gel column chromatography was attempted 25.1, 24.3, 22.8, 14.2; HRMS (ESI m/z): calcd for [C33H49N2O6]+
for analysis purposes with 80% EtOAc–20% petroleum spirit 569.3585, found 569.3592; n(max) cmꢀ1: 2923 s, 2853 s, 1698 s,
solution and the resulting light brown oil was shown by 1521 m, 1132 m. Palladium on activated carbon (31 mg, 10% w/w)
1H NMR spectroscopy to be the desired diaminooctane diol 9. was then suspended in methanol (50 mL) and left to stir before
1H NMR (270 MHz, CDCl3): d 7.34 (5H, s), 5.07 (2H, s), 3.73 the addition of the diaminobutane dodecyl acetal (314 mg,
(2H, s), 3.43 (2H, s,), 3.17–3.15 (2H, d, J = 8.1 Hz), 3.04 (1H, s), 0.55 mmol) in methanol (50 mL). The reaction mixture was
2.67 (1H, s), 2.14–2.11 (1H, d, J = 12.0 Hz), 1.73 (1H, s), stirred under H2(g) for 24 h before being vacuum filtrated
c
1902 New J. Chem., 2013, 37, 1895--1905
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Paper
through a celite plug and the filtrate removed in vacuo to give a in the next deprotection step. 1H NMR (270 MHz, CDCl3):
1
white paste. Analysis by H NMR spectroscopy determined the d 7.32 (5H, s), 5.02 (2H, s), 4.59–4.55 (1H, t, J = 10.8 Hz), 3.83
loss of the singlet resonance at d 5.07 ppm indicating the (2H, s), 3.58–3.57 (2H, t, J = 4.0 Hz, 2 ꢁ CH), 3.38–3.36 (2H, d, J =
successful deprotection of the benzyl carbamate to afford 5.4 Hz), 2.96–2.95 (1H, d, J = 2.7 Hz, 1 ꢁ CH), 2.78 (1H, s, 1 ꢁ
compound 13 (211 mg, 88%) in >95% purity which was then CH), 2.33–2.27 (1H, t, J = 8.1 Hz), 1.99–1.95 (1H, d, J = 10.8 Hz),
used without further purification. HRMS (ESI m/z): calcd for 1.58–1.57 (2H, m, CH2), 1.31 (18H, bs), 0.88–0.84 (3H, t, J =
[C25H43N2O4 H+] 435.3217, found 435.3233; n(max) cmꢀ1: 2923 s, 10.8 Hz); 13C NMR (67.5 MHz, CDCl3): d 176.8, 156.6, 136.5,
2853 s, 1697 s.
128.6–128.3, 104.3, 78.4–76.7, 66.9, 44.6, 42.6, 39.8, 38.3, 35.9,
34.0, 32.7, 32.0, 31.10, 29.7–29.5, 24.33, 22.8, 14.3; calcd for
[C31H45N2O6]+ 541.3272, found 541.3301; n(max) cmꢀ1: 3360 m,
2924 s, 2854 s, 1699 s, 1522 m, 1142 m. Deprotection of the
diaminoethane dodecyl acetal (459 mg, 0.85 mmol) was then
Synthesis of benzyl [2-(2hexaodecyl-5,7-dioxooctahydro-6H-4,8-
methano[1,3]dioxolo[4,5-f]isoindol-6-yl)butyl]carbamate and
the corresponding deprotected amphiphiles 14
Synthesis of diaminobutane hexadecyl acetal was performed as carried out as per previously described for diaminobutane
per previously described for diaminobutane dodecyl acetal with dodecyl acetal with palladium on activated carbon (46 mg,
1
diaminobutane diol 8 (246 mg, 0.613 mmol) and hexadecyl 10% w/w) to afford a white paste. Analysis by H NMR spectro-
aldehyde (221 mg, 0.919 mmol) to afford a light yellow oil. scopy determined the loss of the singlet resonance at d 5.02 ppm
Purification by silica gel column chromatography was per- indicating the successful deprotection of the benzyl carbamate
formed with a 70% petroleum spirit–30% EtOAc solution and to afford compound 11 (319 mg, 92%) in >95% purity which
1
the resulting caramel oil was shown by H NMR spectroscopy was then used without further purification. HRMS (ESI m/z):
analysis showed it to be the desired diaminobutane hexadecyl calcd for [C23H39N2O4]+ 407.2911, found 407.2931; n(max) cmꢀ1
:
acetal (286 mg, 75%) in >95% purity which was then used 2954 s, 2853 s, 1699 s.
1
immediately in the next deprotection step. H NMR (270 MHz,
CDCl3): d 7.38–7.28 (5H, m), 5.06 (2H, s), 4.62–4.59 (1H, t, J =
Synthesis of benzyl [2-(2-hexadecyl-5,7-dioxooctahydro-6H-4,8-
5.4 Hz), 3.84 (2H, s), 3.44–3.40 (2H, t, J = 5.4 Hz, 2 ꢁ CH),
methano[1,3]dioxolo[4,5-f]isoindol-6-yl)butyl]carbamate and
3.22–3.17 (2H, d, J = 13.5 Hz), 3.08–3.06 (2H, m, 2 ꢁ CH), 2.83–
the corresponding deprotected amphiphile 12
2.81 (2H, m, 2 ꢁ CH), 2.02–1.97 (2H, d, J = 13.5 Hz), 1.62–1.32
(6H, m, 3 ꢁ CH2), 1.22 (26H, bs), 0.88–0.83 (3H, t, J = 5.4 Hz) Synthesis of diaminoethane hexadecyl acetal 12 was performed
13C NMR (67.5 MHz, CDCl3): d 179.7, 176.4, 156.5, 139.4, 136.5, as per previously described for diaminobutane dodecyl acetal
128.6–128.2, 114.4, 104.1, 77.6–76.6, 66.8, 44.5, 42.7, 40.5, with diaminoethane diol 7 (250 mg, 0.668 mmol) and hexadecyl
38.2, 35.9, 34.2, 32.7, 32.0, 29.8–29.03, 27.5, 25.1, 24.7, 24.3, aldehyde (240 mg, 1.00 mmol) to afford a light yellow oil.
22.8, 14.2; HRMS (ESI m/z): calcd for [C37H57N2O6]+ 625.4211, Purification by silica gel column chromatography was per-
found 625.4200; n(max) cmꢀ1: 3368 m, 2924 s, 2853 s, 1697 s, formed with a 70% petroleum spirit–30% EtOAc solution and
1
1532 m, 1137 m. Deprotection of diaminobutane hexadecyl the resulting caramel oil was shown by H NMR spectroscopy
acetal (377 mg, 0.60 mmol) was then carried out as per analysis showed it to be the desired diaminoethane hexadecyl
previously described for diaminobutane dodecyl acetal with acetal (276 mg, 70%) in >95% purity which was then used
1
palladium on activated carbon (38 mg, 10% w/w) to afford a immediately in the next deprotection step. H NMR (270 MHz,
1
white paste. Analysis by H NMR spectroscopy determined the CDCl3): d 7.32 (5H, s), 5.02 (2H, s), 4.58–4.55 (1H, t, J = 2.7 Hz),
loss of the singlet resonance at d 5.06 ppm indicating the 3.83 (2H, s), 3.60–3.56 (2H, t, J = 5.4 Hz, 2 ꢁ CH), 3.40–3.34 (2H,
successful deprotection of the benzyl carbamate to afford d, J = 5.4 Hz), 2.97–2.93 (2H, m, 2 ꢁ CH), 2.79–2.78 (2H, t, J =
compound 14 (263 mg, 89%) in >95% purity which was then 2.7, 2 ꢁ CH), 2.34–2.28 (1H, t, J = 8.1 Hz), 1.99–1.95 (1H, d,
used without further purification. HRMS (ESI m/z): calcd for J = 10.8 Hz), 1.60–1.57 (2H, m, CH2), 1.34 (26H, bs), 0.88 (3H, t,
[C29H51N2O4]+ 491.3843, found 491.3824; n(max) cmꢀ1: 2920 s, J = 13.5 Hz); 13C NMR (67.5 MHz, CDCl3): d 174.7, 156.6, 136.4,
2851 s, 1698 s.
128.6–128.4, 104.1, 77.7–76.6, 66.9, 44.5, 42.6, 39.6, 38.3, 37.2,
34.0, 32.7, 32.0, 29.8–29.2, 24.8, 24.3, 22.7, 14.21; HRMS (ESI m/z):
calcd for [C35H53N2O6]+ 597.3898, found 597.3912; n(max) cmꢀ1
:
Synthesis of benzyl [2-(2-dodecyl-5,7-dioxooctahydro-6H-4,8-
methano[1,3]dioxolo[4,5-f]isoindol-6-yl)ethyl]carbamate and
the corresponding deprotected amphiphile 11
3354 m, 2922 s, 2853 s, 1698 s, 1526 m, 1120 m. Deprotection of
the diaminoethane hexadecyl acetal (62 mg, 0.10 mmol) was
Synthesis of diaminoethane dodecyl acetal 11 was performed as then carried out as per previously described for diaminobutane
per previously described for diaminobutane dodecyl acetal with dodecyl acetal with palladium on activated carbon (6 mg, 10%
1
diaminoethane diol 7 (212 mg, 0.566 mmol) and dodecyl w/w) to afford a white paste. Analysis by H NMR spectroscopy
aldehyde (0.19 mL, 0.849 mmol) to afford a light yellow oil. determined the loss of the singlet resonance at d 5.02 ppm
Purification by silica gel column chromatography was performed indicating the successful deprotection of the benzyl carbamate
with a 60% petroleum spirit–40% EtOAc solution and the to afford compound 12 (40 mg, 83%) in >95% purity which was
resulting caramel oil was shown by 1H NMR spectroscopy analysis then used without further purification. HRMS (ESI m/z): calcd
showed it to be the desired diaminoethane dodecyl acetal for [C27H47N2O4]+ 463.3530, found 463.3527; n(max) cmꢀ1: 2916 s,
(188 mg, 61%) in >95% purity which was then used immediately 2849 s, 1698 s.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 1895--1905 1903

View Article Online
Paper
NJC
Synthesis of benzyl [2-(2-dodecyl-5,7-dioxooctahydro-6H-4,8-
methano[1,3]dioxolo[4,5-f]isoindol-6-yl)octyl]carbamate and
the corresponding deprotected amphiphiles 15
78.33, 66.69, 44.58, 42.73, 41.14, 38.72, 32.69, 32.11, 29.78, 27.80,
26.67, 24.22, 22.80, 14.22; HRMS (ESI m/z): calcd for [C41H65N2O6]+
681.4837, found 681.4862; n(max) cmꢀ1: 3368 m, 2924 m, 2853 m,
1697 s, 1532 m, 1137 m. Deprotection of the diaminooctane
hexadecyl acetal (95 mg, 0.140 mmol) was then carried out as
per previously described for diaminobutane dodecyl acetal with
palladium on activated carbon (20 mg, 10% w/w) to afford a cream
Synthesis of acetal 15 was performed as per previously described
for diaminobutane dodecyl acetal with diaminooctane diol
9 (135 mg, 0.294 mmol) and dodecyl aldehyde (0.13 mL,
0.589 mmol) to afford a brown oil. Purification by silica gel
column chromatography was performed with a 60% petroleum
spirit–40% EtOAc solution and the resulting caramel oil was
shown by 1H NMR spectroscopy analysis showed it to be the
desired diaminooctane dodecyl acetal (235 mg, 64%) in >95%
purity which was then used immediately in the next deprotection
step. 1H NMR (270 MHz, CDCl3): d 7.34–7.33 (5H, m), 5.07
(2H, s), 4.60–4.58 (1H, t, J = 2.7 Hz), 3.85 (1H, s), 3.42–3.37
(2H, t, J = 8.1 Hz, 2 ꢁ CH), 3.19–3.12 (2H, q, J = 5.4 Hz),
3.08–3.05 (2H, m, 2 ꢁ CH), 2.83 (2H, br s, 2 ꢁ CH), 2.34–2.26
(1H, m, bridge-H) 1.97 (1H, s, bridge-H), 1.65–1.58 (5H, m, 2 ꢁ
CH2) 1.48–1.46 (2H, m, 1 ꢁ CH2), 1.26–1.24 (27H, m), 0.85 (3H, t,
J = 5.4 Hz); 13C NMR (67.5 MHz, CDCl3): d 176.41–176.36, 156.48,
136.75, 128.60–128.54, 128.18, 128.14, 104.14, 78.32, 66.63, 44.48,
42.70, 41.12, 38.68, 35.87, 33.95, 32.69, 31.97, 29.69, 29.67–29.16,
27.76, 26.89, 26.66, 24.87, 24.24, 22.76, 22.75, 22.74, 21.10, 14.18;
HRMS (ESI m/z): calcd for [C37H57N2O6]+ 625.4211, found
625.4224; n(max) cmꢀ1: 3365 m, 2965 m, 2850 m, 1694 s, 1537 m,
1129 m. Deprotection of the diaminooctane dodecyl acetal
(235 mg, 0.375 mmol) was then carried out as per previously
described for diaminobutane dodecyl acetal with palladium on
activated carbon (24 mg, 10% w/w) to afford a cream paste.
Analysis by 1H NMR spectroscopy determined the loss of the
singlet resonance at d 5.02 ppm indicating the successful depro-
tection of the benzyl carbamate to afford compound 15 (147 mg,
89%) in >95% purity which was then used without further
purification. HRMS (ESI m/z): calcd for [C29H51N2O4]+ 491.3843,
found 491.3861; n(max) cmꢀ1: 2927 s, 2850 s, 1699 s.
1
paste. Analysis by H NMR spectroscopy determined the loss of
the singlet resonance at d 5.02 ppm indicating the successful
deprotection of the benzyl carbamate to afford compound 16
(80 mg, 84%) in >95% purity which was then used without further
purification. HRMS (ESI m/z): calcd for [C33H59N2O4]+ 547.4469,
found 547.4476; n(max) cmꢀ1: 2920 s, 2851 s, 1699 s.
Synthesis of benzyl [2-(2-dodecyl-5,7-dioxooctahydro-6H-4,8-
methano[1,3]dioxolo[4,5-f]isoindol-6-yl)decyl]carbamate and
the corresponding deprotected amphiphile 17
Synthesis of diaminodecane dodecyl acetal 17 was performed as
per previously described for diaminobutane dodecyl acetal with
diaminodecane diol 10 (361 mg, 0.742 mmol) and dodecyl
aldehyde (0.25 mL, 1.11 mmol) to afford a brown oil. Purifica-
tion by silica gel column chromatography was performed with a
60% petroleum spirit–40% EtOAc solution and the resulting
caramel oil was shown by 1H NMR spectroscopy analysis
showed it to be the desired diaminodecane dodecyl acetal
(368 mg, 76%) in >95% purity which was then used immedi-
ately in the next deprotection step. 1H NMR (270 MHz, CDCl3):
d 7.33 (5H, br s), 5.08 (2H, s), 4.61–4.57 (1H, t, J = 5.4 Hz), 3.85
(3H, br s), 3.43–3.37 (1H, t, J = 8.1 Hz), 3.18–3.15 (1H, d, J =
8.1 Hz), 3.09–3.06 (2H, q, J = 5.4 Hz) 2.83 (2H, br s), 2.36–2.20
(1H, m) 2.01–1.97 (1H, d, J = 10.8 Hz), 1.64–1.59 (7H, m), 1.24
(35H, m), 0.88–0.84 (3H, t, J = 5.4 Hz); d 13C NMR (67.5 MHz,
CDCl3): 174.32, 156.50, 136.73, 128.59–128.22, 104.34, 78.33,
66.67, 44.48, 42.71, 38.79, 35.89, 34.06, 32.69, 31.99, 30.00–
29.09, 27.02, 24.78, 24.26, 22.77, 14.20; HRMS (ESI m/z): calcd
for [C39H61N2O6]+ 653.4524, found 653.4543; n(max) cmꢀ1: 3349 m,
2922 m, 2870 m, 1695 s, 1538 m, 1142 m. Deprotection of the
diaminodecane dodecyl acetal (368 mg, 0.564 mmol) was then
carried out as per previously described for diaminobutane dodecyl
Synthesis of benzyl [2-(2hexadecyl-5,7-dioxooctahydro-6H-4,8-
methano[1,3]dioxolo[4,5-f]isoindol-6-yl)octyl]carbamate and
the corresponding deprotected amphiphiles 16
Synthesis of diaminooctane hexadecyl acetal 16 was performed acetal with palladium on activated carbon (37 mg, 10% w/w) to
1
as per previously described for diaminobutane dodecyl acetal afford a cream paste. Analysis by H NMR spectroscopy deter-
with diaminooctane diol 9 (164 mg, 0.358 mmol) and hexadecyl mined the loss of the singlet resonance at d 5.02 ppm indicating
aldehyde (129 mg, 0.537 mmol) to afford a brown oil. Purifica- the successful deprotection of the benzyl carbamate to afford
tion by silica gel column chromatography was performed with a compound 17 (234 mg, 80%) in >95% purity which was then
60% petroleum spirit–40% EtOAc solution and the resulting used without further purification. HRMS (ESI m/z): calcd for
caramel oil was shown by 1H NMR spectroscopy analysis [C31H56N2O4]+ 463.3537, found 463.3627; n(max) cmꢀ1: 2922 s,
showed it to be the desired diaminooctane hexadecyl acetal 2870 s, 1699 s.
(95 mg, 39%) in >95% purity which was then used immediately
in the next deprotection step. 1H NMR (270 MHz, CDCl3):
Synthesis of benzyl [2-(2-hexaecyl-5,7-dioxooctahydro-6H-4,8-
methano[1,3]dioxolo[4,5-f]isoindol-6-yl)decyl]carbamate and
the corresponding deprotected amphiphiles 18
d 7.35–7.33 (5H, m), 5.12 (2H, s), 4.60–4.58 (1H, t, J = 2.7 Hz),
3.85 (1H, s), 3.40–3.37 (2H, t, J = 8.1 Hz, 2 ꢁ CH), 3.20–3.12
(2H, q, J = 5.4 Hz), 3.07–3.06 (2H, m, 2 ꢁ CH), 2.88–2.83 (4H, m, Synthesis of diaminodecane hexadecyl acetal 18 was performed
2 ꢁ CH), 2.35–2.29 (1H, t, J = 8.1 Hz) 2.01–1.97 (1H, d, J = 10.8 Hz), as per previously described for diaminobutane dodecyl acetal
1.48–1.46 (6H, m, 3 ꢁ CH2), 1.40–1.35 (6H, m, 3 ꢁ CH2) with diaminodecane diol 10 (132 mg, 0.271 mmol) and hexa-
1.23–1.20 (30H, m), 0.87 (3H, t, J = 2.7 Hz, CH3); 13C NMR decyl aldehyde (98 mg, 0.406 mmol) to afford a brown oil.
(67.5 MHz, CDCl3): 176.37, 156.47, 136.74, 128.89–128.20, 104.08, Purification by silica gel column chromatography was performed
c
1904 New J. Chem., 2013, 37, 1895--1905
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Paper
7 R. Bordes, M. Vedrenne, Y. Coppel, S. Franceschi, E. Perez
and I. Rico-Lattes, ChemPhysChem, 2007, 8, 2013.
8 L. C. Henderson, J. Li, R. L. Nation, T. Velkov and
F. M. Pfeffer, Chem. Commun., 2010, 46, 3197.
9 G. J. Atwell and W. A. Denny, Synthesis, 1984, 1032.
10 P. G. M. Wuts and T. W. Greene, Greene’s Protective Groups in
Organic Synthesis, John Wiley & Sons, Brisbane, 4th edn, 2000.
11 V. K. Ahluwalia, Laboratory Techniques In Organic Chemistry,
I. K. International Publishing House Pvt. Limited, 2005.
12 I. V. Tetko, J. Gasteiger, R. Todeschini, A. Mauri,
D. Livingstone, P. Ertl, V. Palyulin, E. Radchenko, N. S.
Zefirov, A. S. Makarenko, V. Y. Tanchuk and V. V. Prokopenko,
J. Comput.-Aided Mol. Des., 2005, 19, 453.
with a 70% petroleum spirit–30% EtOAc solution and the
resulting caramel oil was shown by 1H NMR spectroscopy
analysis showed it to be the desired diaminodecane hexadecyl
acetal (165 mg, 88%) in >95% purity which was then used
1
immediately in the next deprotection step. H NMR (270 MHz,
CDCl3): d 7.34–7.33 (5H, m), 5.08 (2H, s), 4.59–4.57 (1H, t, J =
2.7 Hz), 3.85 (3H, br s), 3.43–3.37 (1H, t, J = 8.1 Hz, 1 ꢁ CH),
3.20–3.13 (1H, q, J = 5.4 Hz, CH), 3.08–3.05 (2H, q, J = 2.7 Hz, 2 ꢁ
CH), 2.83 (2H, br s, 2 ꢁ CH), 2.34–2.29 (1H, m) 2.01–1.97 (1H, d,
J = 10.8 Hz), 1.60–1.36 (7H, m, 3 ꢁ CH2), 1.27–1.19 (39H, m), 0.85
(3H, t, J = 5.4 Hz) d 13C NMR (67.5 MHz, CDCl3): d 176.42, 156.54,
136.76, 128.45–128.00, 104.03, 78.26, 66.50, 44.40, 42.64,
38.65–38.59, 35.76, 34.02, 32.63, 31.95- 31.93, 27.74–27.70,
26.92–26.69, 25.78, 25.34, 24.77–24.17, 22.72–22.33, 20.93,
14.14; HRMS (ESI m/z): calcd for [C43H68N2NaO6]+ 731.4969,
731.4952; found n(max) cmꢀ1: 3359 m, 2923 m, 2853 m, 1699 s,
1558 m, 1144 m. Deprotection of the diaminodecane hexadecyl
acetal (46 mg, 0.066 mmol) was then carried out as per pre-
viously described for diaminobutane dodecyl acetal with palla-
dium on activated carbon (4.6 mg, 10% w/w) to afford a cream
13 VCCLAB, Virtual Computational Chemistry Laboratory,
http://www.vcclab.org, 2005.
14 (a) A. Roda, A. Minutello, M. A. Angellotti and A. Finit,
J. Lipid Res., 1990, 31, 1433–1443; (b) M. Abla, G. Durand
and B. Pucci, J. Org. Chem., 2008, 73(21), 8142–8153;
(c) H. Hong, L. Wang and G. Zou, J. Liq. Chromatogr. Relat.
Technol., 1997, 20(18), 3029; (d) M. Grovera, M. Gulatia,
B. Singh and S. R. Singh, QSAR Comb. Sci., 2005, 24, 639.
15 S. O. Kelley, K. M. Stewart and R. Mourtada, Pharm. Res.,
2011, 28, 2808–2819.
16 T. Asakawa, M. Mouri, S. Miyagishi and M. Nishida, Langmuir,
1989, 5, 343.
17 Y. Shi, H. Q. Luo and N. B. Li, Spectrochim. Acta, Part A, 2011,
78, 1403.
1
paste. Analysis by H NMR spectroscopy determined the loss of
the singlet resonance at d 5.02 ppm indicating the successful
deprotection of the benzyl carbamate to afford compound 18
(26 mg, 67%) in >95% purity which was then used without
further purification. HRMS (ESI m/z): calcd for [C35H62N2O4H]+
575.4789, found 575.4863; n(max) cmꢀ1: 2920 s, 2851 s, 1698 s.
18 Due to the sparingly soluble nature of these compounds in
water, it is unlikely that a concentration of 1 mg mLꢀ1 was
achieved. In essence a 1 mL water sample was saturated
with compound, this is markedly higher than the CAC value
and thus represents no experimental error.
19 T. Kunitake, Y. Okahata, M. Shimomura, S. I. Yasunami and
K. Takarabe, J. Am. Chem. Soc., 1981, 103, 5401.
20 T. H. Zhang and R. E. Marchant, J. Colloid Interface Sci.,
1996, 177, 419.
Acknowledgements
The authors would like to thank the Deakin University Central
Research Grants Scheme, The Institute for Frontier Materials,
The CASS foundation and the Strategic Research Centre for
Chemistry and Biotechnology for funding. We would also like
to thank the Australian Government for the Australian Post-
graduate Award for J.S. Thanks are due to Dr Pierre Guillet
(University of Avignon) for discussions on DLS and TEM data.
21 K. Shinoda, M. Hato and T. Hayashi, J. Phys. Chem., 1972,
76, 909.
22 V. Tomasic, A. Tomasic, I. Smit and N. Filipovic-Vincekovic,
J. Colloid Interface Sci., 2005, 285, 342.
Notes and references
1 (a) J. Buse and A. El-Aneed, Nanomedicine, 2010, 5, 1237;
(b) C. Byrne, F. Sallas, D. K. Rai, J. Jogier and R. Darcy, Org. 23 P. Mukerjee, K. J. Mysels, Critical micelle concentrations of
Biomol. Chem., 2009, 7, 3763; (c) R. K. Tekade, P. V. Kumar
and N. K. Jain, Chem. Rev., 2009, 109, 49; (d) G. Sahay,
aqueous surfactant systems; U.S. National Bureau of Standards;
for sale by the Supt. of Docs., U.S. Govt. Print Off, 1971.
D. Y. Alakhova and A. V. Kabanov, J. Controlled Release, 24 D. Mitra and S. P. Moulik, J. Chem. Sci., 2010, 122, 349.
2010, 145, 182; (e) Z. Gao, L. Zhang and Y. Sun, J. Controlled 25 S. Bhattacharya and S. N. G. Acharya, Langmuir, 2000, 16, 87.
Release, 2012, 162, 45.
26 F. M. Menger and L. Shi, J. Am. Chem. Soc., 2006, 128,
2 (a) H. Kaya-Celiker and K. Mallikarjunan, Food Eng. Rev., 2012,
9338–9339.
4, 114; (b) M. J. Joralemon, S. McRae and T. Emrick, Chem. 27 D. Lundberg, L. Shi and F. M. Menger, Langmuir, 2008, 24,
Commun., 2010, 46, 1377; (c) H. Maeda, J. Wu, T. Sawa, 4530–4536.
Y. Matsumura and K. Hori, J. Controlled Release, 2000, 65, 271. 28 S. Bhattacharya and J. Biswas, Langmuir, 2011, 27, 1581.
3 M. A. Mintzer and E. E. Simanek, Chem. Rev., 2009, 109, 259. 29 J. N. Israelachvili, D. J. Mitchell and B. W. Ninham, Biochim.
4 S. Bhattacharya and A. Bajaj, Chem. Commun., 2009, 4632.
5 T. Barclay, K. Constantopoulos and J. Matisons, J. Mater. 30 T. Shimizu, M. Masuda and H. Minamikawa, Chem. Rev.,
Res., 2011, 26(2), 322. 2005, 105, 401.
6 R. Bordes, K. Rbii, A. Gonzalez-Perez, S. Franceschi-Messant, 31 J. N. Israelachvili, D. J. Mitchell and B. W. Ninham, J. Chem.
Biophys. Acta, Biomembr., 1977, 470, 185.
E. Perez and I. Rico-Lattes, Langmuir, 2007, 23, 7526.
Soc., Faraday Trans. 2, 1976, 72, 1525.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 1895--1905 1905