Synthesis and properties of multiarmed geminis

Article abstract of DOI:10.1021/jo9912350

Three families of multiarmed and multicationic gemini surfactants, having pentaerythritol, dipentaerythritol, or adamantane cores, were synthesized and examined for their colloidal properties in water. Geminis with four or six six-carbon chains are highly water-soluble and do not self- assemble. But surface tension and conductivity measurements show that the geminis with four or six eight-carbon chains form micelles in the 3-6 mM range (compared to 0.5 M for a corresponding surfactant with a single chain). According to dynamic light scattering, these micelles are small (<30 A diameter); no evidence of dendritic growth below 25 mM is evident. Geminis with four or six 12-carbon chains are too water-insoluble to examine for micelle formation. It is concluded the outward projection of the hydrocarbon chains in water greatly enhances the propensity of the surfactants to self- assemble. Micellar growth is seemingly restricted by chain pairing, chain looping, and associative ring formation. Since the aggregates have, despite these effects, greater residual water-hydrocarbon contact than found in typical micelles, water solubility of surfactants having longer chains is impaired.

Full text of DOI:10.1021/jo9912350

8916
J . Org. Chem. 1999, 64, 8916-8921
Syn th esis a n d P r op er ties of Mu ltia r m ed Gem in is
Fredric M. Menger* and Vasily A. Migulin
Department of Chemistry, Emory University, Atlanta, Georgia 30322
Received August 2, 1999
Three families of multiarmed and multicationic gemini surfactants, having pentaerythritol,
dipentaerythritol, or adamantane cores, were synthesized and examined for their colloidal properties
in water. Geminis with four or six six-carbon chains are highly water-soluble and do not self-
assemble. But surface tension and conductivity measurements show that the geminis with four or
six eight-carbon chains form micelles in the 3-6 mM range (compared to 0.5 M for a corresponding
surfactant with a single chain). According to dynamic light scattering, these micelles are small
(<30 Å diameter); no evidence of dendritic growth below 25 mM is evident. Geminis with four or
six 12-carbon chains are too water-insoluble to examine for micelle formation. It is concluded the
outward projection of the hydrocarbon chains in water greatly enhances the propensity of the
surfactants to self-assemble. Micellar growth is seemingly restricted by chain pairing, chain looping,
and associative ring formation. Since the aggregates have, despite these effects, greater residual
water-hydrocarbon contact than found in typical micelles, water solubility of surfactants having
longer chains is impaired.
In tr od u ction
Scheme 1 gives three examples with labels that include
a number indicating the chain length of the tails. Do
these compounds abruptly self-assemble at a critical
micelle concentration (CMC)? If so, how big are the
micelles? And how do the surfactants affect the surface
tension of water? Such questions are addressed after first
discussing how the compounds were synthesized.
Kary Mullis, a Nobel laureate of “polymerase chain
reaction” fame, wrote recently: “Probably the most
important scientific development of the twentieth century
is that economics has replaced curiosity as the driving
force behind research”.1 As with many such generaliza-
tions, the message seems overblown. Perhaps a more
accurate statement would be that economics has increas-
ingly joined (but, fortunately, not yet replaced) curiosity
as a major driving force. Gemini surfactants, having a
structure shown in Chart 1, are a case in point. We were
first prompted to synthesize geminis (and, indeed, to coin
the name) out of curiosity over how preventing intramo-
lecular chain-chain association via long, rigid spacers
(e.g., a stilbene unit) would affect self-assembly in
water.2,3 Later on, many other investigators, focusing on
more flexible spacers, began patenting geminis for their
potential commercial utility.4 Economic considerations
had joined curiosity to generate a vibrant field of re-
search. Although the field is still young, short reviews of
the progress to date are already available.4-6
Syn th esis
The synthesis of the PE, DPE, and AD surfactants is
described in Scheme 2, parts A, B, and C, respectively.
More straightforward attempts to attach long-chain
tertiary amines to the PE core via the tetrabromo
(Aldrich), tetramesylate,8 and tetratriflate derivatives of
pentaerythritol failed, thus accounting for the tetraester
approach seen in Scheme 2. Since spontaneous ester
hydrolysis was a concern, we invariably checked our
geminis by NMR for this potential problem. It turns out
that PE-8, for example, exhibited no hydrolysis after
standing in D2O for 24 h at room temperature and 4 h
at 55 °C. Perhaps steric hindrance, arising from bulky
groups on both sides of the ester, stabilizes the com-
pounds. Of the compounds examined, only PE-12 and
DPE-12 hydrolyzed slowly, but their water solubility was
too low for practical purposes anyway. Curiously, PE-8
and other geminis are much more labile in CD3OD,
transesterifying in less than 7 h at room temperature;
thus, alcoholic solvents were always avoided in purifica-
tion schemes.
It was a simple extension of our work and that of
others7 to wonder about the behavior of surfactants with
four or more hydrocarbon tails (which we loosely also
include in the gemini family). This led to our synthesizing
three series of compounds based on pentaerythritol (PE),
dipentaerythritol (DPE), and adamantane (AD). Each
possesses multiple water-solubilizing quaternary nitro-
gens bearing hydrocarbon chains of varying length.
(1) Mullis, K. Dancing Naked in the Mind Fields; Panthem Books:
New York, 1998.
(2) Menger, F. M.; Littau, C. A. J . Am. Chem. Soc. 1991, 113, 1451.
(3) Menger, F. M.; Littau, C. A. J . Am. Chem. Soc. 1993, 115, 10083.
(4) Rosen, M. J .; Tracy, D. J . J . Surfactants Deterg. 1998, 1, 52/7.
(5) Rosen, M. J . CHEMTECH 1993, 23, 30.
(6) Menger, F. M.; Keiper, J . S. Review to be published in Angew.
Chem. This review covers a history of geminis, including their early
synthesis, as well as a multitude of papers published from all over
the world.
Synthetic details are given in the Experimental Sec-
tion. Recrystallization or multiple washings with acetone
gave geminis as white solids of correct structure and
1
satisfactory purity as judged by H and 13C NMR and by
elemental analysis. Yields ranged from good to excellent,
although no attempt was made to optimize them.
(7) Sumida, Y.; Masuyama, A.; Maekawa, H.; Takasu, M.; Kida, T.;
Nakatsuji, Y.; Ikeda, I.; Nojima, M. Chem. Commun. 1998, 2385. See
also refs 4-6.
(8) Turkevich, B. M. Farm. Zh. (Kiev) 1962, 17, 14; Chem. Abstr.
1963, 59, 1471.
10.1021/jo9912350 CCC: $18.00 © 1999 American Chemical Society
Published on Web 11/09/1999

Synthesis and Properties of Multiarmed Geminis
J . Org. Chem., Vol. 64, No. 24, 1999 8917
Ch a r t 1
Sch em e 1
Sch em e 2
Ta ble 1. Solu bilities of Gem in is in Ch lor ofor m a n d
Wa ter a
reveal overall “hydrophobicites” and because they allow
others to better plan their experiments. Table 1 shows a
high chloroform solubility (i.e., 0.3 M or higher) for all
PEs and DPEs investigated (even those with only six-
carbon chains). Except for AD-6, the AD solubility in
chloroform is in the mM range. Whether or not reverse
micelles are formed in chloroform was not determined.
gemini
CHCl3
H2O
PE-6
PE-8
>0.5 M
>0.5 M
>0.5 M
30-50 mM
<0.2 mM
>0.5 M
PE-12
DPE-6
DPE-8
DPE-12
AD-6
AD-8
AD-10
AD-12
>0.5 M
0.3-0.5 M
>0.5 M
0.3-0.5 M
<0.1 mM
0.5-1 mM
1-3 mM
3-5 mM
50-100 mM
<0.1 mM
>0.5 M
50-80 mM
2-3 mM
<0.1 mM
Water solubility at room temperature depends mark-
edly upon chain length. Thus, PE-6, DPE-6, and AD-6
are all highly water-soluble (>0.5 M). In contrast, PE-
12, DPE-12, and AD-12 are all rather water insoluble
(<0.2 mM). This extreme sensitivity to chain length
cannot be ascribed simply to “multiple chains” because
the geminis all have one cationic charge per chain, which
is no different from that in water-soluble conventional
surfactants such as dodecyltrimethylammonium bromide.
Clearly, geometry and connectivity are of the essence. We
presume that our multiarmed geminis have difficulty
self-assembling into aggregates where each chain can
immerse itself in a hydrophobic environment (as occurs
with conventional surfactants when they form spherical
micelles). The point is clarified in Figure 1 in which it is
a
At 23 °C in units of molarity and micromolarity.
P r op er ties
Solu bility. Solubility is a surprisingly difficult pa-
rameter to determine accurately.9 In our case, we were
content to define only solubility ranges in chloroform and
water (Table 1). Although often omitted from descriptions
of surfactants, solubilities are important because they
(9) Bohon, R. L.; Claussen, W. F. J . Am. Chem. Soc. 1951, 73, 1571.

8918 J . Org. Chem., Vol. 64, No. 24, 1999
Menger and Migulin
F igu r e 2. Surface tension of C6-chained geminis.
F igu r e 1. Schematic of the first steps in a possible “dendritic
growth” of a multiarmed gemini.
seen that hydrophobic association between two ADs, each
contributing one chain, leaves six exposed chains. If each
of these six “loose ends” similarly associated with another
gemini, then a “dendritic-like” aggregate with 18 exposed
hydrocarbon chains would ensue. Further growth could,
in principle, develop into a gelating network (a possibility
that prompted, in part, this work). But, as will be shown,
this did not occur. The geminis prefer to self-assemble
into small aggregates or, alternatively, to resist dissolving
in water altogether despite the multiple ionic charges.
In addition to the enforced dispositions of the hydro-
carbon chains among the geminis, the surfactants have
interconnected ionic groups. This too affects solubility
adversely as one would not expect, for example, four
interconnected cations to solubilize four 12-carbon sub-
stituents as effectively as four independent cations solu-
bilize one 12-carbon chain each.
F igu r e 3. Surface tension of DPE-C8 water solutions.
Gemini AD-10, whose modest water solubility of 3 mM
increases 3-fold as the temperature is increased from 20
to 40 °C, shows no obvious Krafft point within this range.
Solubility, of course, depends on solid-state forces as well
as solution behavior. We are currently attempting to
elucidate the solid-state properties of the geminis, includ-
ing their tendency to form fibers, a topic upon that we
will amplify at a later date.
Micelle F or m a tion . Micelle formation is most com-
monly detected by plotting surface tension vs log [sur-
factant]. The surface tension will decrease until a point
is reached where the plot abruptly levels off; this point
is taken to be the CMC. The sharp bend in the plot
results from surfactant joining micelles rather than
adsorbing at the air/water interface.
Neither PE-6, DPE-6, nor AD-6 forms micelles accord-
ing to surface tension data collected from 0.1 to 25 mM.
For example, as seen in Figure 2, the surface tension of
PE-6 diminishes from 73 to only 57 dynes/cm at 25 mM,
and no break in the plot is evident. A lack of micelle
formation is not surprising, perhaps, because a six-carbon
chain is too short to induce aggregation of conventional
surfactants below 1 M.
F igu r e 4. Specific conductance of DPE-C8 water solutions.
tension vs log [DPE-8] plot (Figure 3) where the surface
tension falls steeply to 40 dynes/cm whereupon it levels
off owing to micelle formation. The break point corre-
sponds to a CMC of 3 mM. Surface tension-based CMC
values for PE-8 and AD-8 are 5 and 6 mM, respectively.
These numbers lie far below the CMC of 0.5 M for
octyltrimethylammonium bromide.10 A plot of specific
conductance vs [DPE-8] in Figure 4 has a break at 3.9
mM, thereby confirming the presence of micelles as
suggested by the surface tension data.
The multiarmed geminis with 12 carbons per chain are
too insoluble (Table 1) to allow meaningful surface
tension measurements. Moreover, it was not possible to
make water-supported monomolecular films with them
because their water solubility, low as it is, was sufficient
to destabilize the films when compressed.
The situation is quite different with the three families
of geminis having four or six eight-carbon chains. The
behavior of DPE-8, for example, is portrayed in a surface
(10) Zana, R. J . Colloid Interface Sci. 1980, 78, 330.

Synthesis and Properties of Multiarmed Geminis
J . Org. Chem., Vol. 64, No. 24, 1999 8919
particularly sensitive to chain length as indicated, for
example, by the water insolubility of geminis with 12-
carbon chains.
It seems clear that in order to achieve larger micelles,
and possibly even a surfactant network, we must impart
to the surfactants a more water-solubilizing functionality
than a quaternary ammonium nitrogen. Work is this
direction is underway.
Exp er im en ta l Section
Gen er a l Meth od s. Melting points are uncorrected. All
reactions were carried out under a nitrogen atmosphere.
Solvents were reagent grade and in most cases dried prior to
use. All reagents were purchased from Aldrich. Amines were
redistilled prior to use. Column chromatography was per-
formed on silica gel 60 (Merk 230-400 mesh). 1H NMR spectra
were recorded at 400 MHz, and 13C NMR spectra were
recorded at 100 MHz. Tensiometry was performed using 2 mL
of freshly prepared solutions with a Kibron Microtrough S
apparatus. Particle-sizing was accomplished utilizing a Coulter
N4 Plus instrument. Conductivity measurements were carried
out with the aid of a YSI model 35 digital conductance meter.
F igu r e 5. Highly schematic representation of a multiarmed
gemini micelle showing ring formation, chain pairing, and
chain coiling, all of which serve to diminish dendritic growth.
(Light scattering data require only a few molecules per
micelle.)
Dynamic light scattering experiments on PE-8, DPE-
8, and AD-8, performed above their CMCs at 25 mM,
provided information on the micelles’ hydrodynamic
diameters. In all three cases, the aggregates were small
(<30 Å, the resolution limit of our instrumentation). We
prefer to call the aggregates “micelles”, but this is
semantic matter. In any event, there is no evidence for
dendritic growth of the surfactants.
P en ta er yth r itol Tetr a tr ifla te. A stirred suspension of
pentaerythritol (0.5 g, 3.68 mmol) in 2.5 mL of pyridine and 4
mL of CH3CN was cooled to 0 °C. Triflic anhydride (5 g, 17.73
mmol) was added dropwise at this temperature over 1 h. The
stirring was continued at rt for 3 h when 1 M HCl (20 mL)
was added. The solution was cooled in an ice bath to complete
the precipitation. The solid was filtered and washed with
water. Recrystallization from CH3CN/H2O gave 2.3 g (94%) of
white crystals: mp ) 165-166 °C; 1H NMR (acetone-d6) δ 5.15
(s, 8H); 13C NMR (acetone-d6) δ 119.5 (q, J ) 317.1 Hz), 72.46,
45.66. Anal. Calcd for C9H8F12O12S4: C, 16.27; H, 1.21; S, 19.30.
Found: C, 16.17; H, 1.12; S, 19.41.
In conclusion, the multiarmed geminis with eight-
carbon chains have a high propensity to aggregate (i.e.,
to have a low CMC) relative to conventional surfactants
of the same chain length. But the micellar aggregates
are small, consisting of only a few molecules as opposed
to the 50 or more that are commonly found in micelles of
single-chain surfactants. And when the chains are mod-
erately long (12 carbons), then water insolubility becomes
an issue. The results point, therefore, to a family of
surfactants that are inordinately hydrophobic considering
that they all possess one cationic charge for each chain.
No doubt this can be attributed to the geometric con-
straints imposed upon the chains. Rather than the chains
being able to direct themselves inwardly into one central
core, as with conventional micelles, the chains are
projected outwardly into the water. Three factors dimin-
ish the likelihood of dendritic growth as pictured in
Figure 1: (a) Molecular models suggest that, even in the
adamantane series, two or three chains within a single
molecule can lie, at least partially, side by side. This is
true for eight-carbon chains; longer chains permit, of
course, even greater van der Waals contact. Such an
intramolecular chain-chain association is evident from
molecular models assuming the preferred s-trans ester
configuration and no more than two gauche linkages in
the hydrocarbon chains. Thus, the geminis are less
“multidentate” than might be presumed. (b) The hydro-
phobic chains can coil and curl, rather than project
linearly into the water, thereby minimizing water-
hydrocarbon contact. In this manner, dendritic growth
would be impaired. (c) A third factor tending to diminish
micellar growth is ring formation. Ring formation is
portrayed in Figure 5 along with intramolecular chain-
chain association and chain coiling, which, as just
discussed, are also likely present in the assemblies. Since
there exists, despite the three factors, a high degree of
residual water-hydrocarbon contact in the gemini ag-
gregates, relative to “normal” micelles, the geminis are
P en ta er yth r itol Tetr a k is(br om oa ceta te). A stirred sus-
pension of pentaerythritol (0.43 g, 3.16 mmol) in 1.02 mL (12.6
mmol) of pyridine and 4 mL of CH3CN was cooled to 0 °C.
BrCH2COBr (1.37 mL, 15.7 mmol) was added dropwise at this
temperature over 1 h. The ice bath was removed, and the
stirring was continued at rt for 5 h. HCl (6 M, 2 mL) was then
added to the reaction mixture together with 3 mL of CH2Cl2.
The aqueous layer was extracted with CH2Cl2 (3 × 5 mL), and
the combined organic layers were concentrated to give a
yellow-orange oil. The oil was dissolved in 5 mL of CH2Cl2 and
washed with brine and with 5 mL of aqueous 0.5 M Na2CO3,
after which the organic layer turned cherry-black. This was
concentrated and purified by filtering through a silica gel
column. Elution with CH2Cl2 yielded 1.59 g (81%) of a yellow
oily solid: 1H NMR (CDCl3) δ 4.28 (s, 8H), 3.86 (s, 8H); 13C
NMR (CDCl3) δ 166.8, 63.5, 42.9, 25.4. Anal. Calcd for C13H16
-
Br4O8: C, 25.19; H, 2.60; Br, 51.56. Found: C, 25.45; H, 2.68;
Br, 51.28.
Dip en ta er yth r itol Hexa k is(br om oa ceta te). A stirred
suspension of dipentaerythritol (0.58 g, 2.3 mmol) in 1.13 mL
(14 mmol) of pyridine and 5 mL of CH3CN was cooled to 0 °C.
BrCH2COBr (2 mL, 22.9 mmol) was added dropwise at this
temperature over 1 h. The ice bath was removed, and stirring
was continued at rt for 5 h. HCl (6 M, 3 mL) was then added
to the reaction mixture together with 3 mL of CH2Cl2. The
aqueous layer was extracted with CH2Cl2 (3 × 5 mL), and the
combined organic layers were concentrated to give a yellow-
orange oil. The oil was dissolved in 5 mL of CH2Cl2 and washed
with brine and with 5 mL of aqueous 0.5 M Na2CO3, after
which the organic layer turned cherry-black. This was con-
centrated and purified by filtering through a silica gel column.
Elution with CH2Cl2 yielded 1.45 g (65%) of a yellow oil: 1H
NMR (CDCl3) δ 4.22 (s, 12H), 3.86 (s, 12H), 3.46 (s, 4H); 13C
NMR (CDCl3) δ 166.8, 68.9, 63.6, 43.7, 25.6. Anal. Calcd for
C
22H28Br6O13: C, 26.97; H, 2.88; Br, 48.93. Found: C, 27.18;
H, 2.90; Br, 48.80.

8920 J . Org. Chem., Vol. 64, No. 24, 1999
Menger and Migulin
1,3,5,7-Tetr a br om oa d a m a n ta n e. The tetrabromide was
prepared by modification of a published procedure.11 Adaman-
tane (9.72 g, 71.5 mmol) was added in small portions over 30
min to a stirred mixture of bromine (45 mL) and anhydrous
aluminum chloride (10 g, 75 mmol) at 0 °C. The mixture was
then slowly heated to 70 °C and held at that temperature for
24 h. Hydrogen bromide evolved vigorously during the addition
and heating. The reaction mixture was treated subsequently
with aqueous sodium sulfite and hydrochloric acid. The
resulting solid was filtered, dried in vacuo over P2O5, and
recrystallized from CH3CN to give 21.2 g (64%) of tan
crystals: mp ) 246-247 °C (lit.11 mp ) 245-247 °C); 1H NMR
(CDCl3) δ 2.71 (s, 12H); 13C NMR (CDCl3) δ 55.0, 54.8. Anal.
Calcd for C10H12Br4: C, 26.58; H, 2.68; Br, 70.74. Found: C,
26.62; H, 2.66; Br, 70.67.
P en ta er yth r itol tetr a k is(d im eth yld od ecyla m m on io-
a ceta te), tetr a br om id e: yield 74%; mp ) 154 °C dec; 1H
NMR (CDCl3) δ 5.36 (s, 8H), 4.35 (s, 8H), 3.75 (t, 8H, J ) 8.4
Hz), 3.50 (s, 24H), 1.78 (m, 8H), 1.38-1.23 (m, 72H), 0.87 (t,
12H, J ) 6.8 Hz); 13C NMR (CDCl3) δ 164.7, 66.4, 64.9, 62.6,
51.0, 41.1, 32.0, 29.7, 29.6, 29.6, 29.5, 29.5, 29.4, 26.4, 22.9,
22.8, 14.3. Anal. Calcd for C69H140Br4N4O8: C, 56.24; H, 9.58;
N, 3.80; Br, 21.69. Found: C, 56.25; H, 9.70; N, 3.76; Br, 21.81.
P en t a er yt h r it ol
Tet r a k is(d im et h yloct yla m m on io-
a ceta te), Tetr a br om id e. In a round-bottom flask, 0.52 g (0.84
mmol) of pentaerythritol tetrakis(bromoacetate) was dissolved
in 10 mL of CHCl3. To the clear solution was added N,N-
dimethyloctylamine (1.03 mL, 5 mmol) in one portion, and the
solution was refluxed for 3 h. CHCl3 was then evaporated, and
the light-brown residue was washed with acetone yielding 0.53
g (51%) of white solid: mp ) 104-105 °C; 1H NMR (CDCl3) δ
5.36 (s, 8H), 4.35 (s, 8H), 3.76 (t, 8H, J ) 8.4 Hz), 3.50 (s, 24H),
1.78 (m, 8H), 1.35-1.25 (m, 40H), 0.86 (t, 12H, J ) 6.8 Hz);
13C NMR (CDCl3) δ 164.7, 66.5, 65.0, 62.6, 51.0, 41.1, 31.8,
1,3,5,7-Tetr a h yd r oxya d a m a n ta n e. The tetraalcohol was
prepared by modification of a published procedure.12 1,3,5,7-
Tetrabromoadamantane (4.17 g, 9.22 mmol) and Ag2SO4 (6.38
g, 20.45 mmol) were suspended in 10 mL of concentrated
sulfuric acid. The mixture was slowly heated to 80 °C and
stirred at this temperature for 5 h. After cooling, the AgBr
precipitate was removed by filtration and washed with water.
The filtrate was neutralized with KOH and evaporated. The
resulting gray residue was dried over P2O5 and then extracted
with 150 mL of ethanol in a Soxhlet apparatus for 12 h. After
removal of ethanol, the residue was dissolved in methanol and
filtered through a short Celite plug. Recrystallization from the
MeOH/EtOH/acetone system afforded 1.5 g (81%) of a white
29.3, 29.2, 26.4, 22.9, 22.7, 14.2. Anal. Calcd for C53H108
-
Br4N4O8: C, 50.96; H, 8.72; N, 4.49; Br, 25.59. Found: C, 51.16;
H, 8.62; N, 4.44; Br, 25.40.
Gen er a l P r oced u r e for th e P r ep a r a tion of th e Gem in i
Su r fa cta n ts Ba sed on th e Dip en ta er yth r itol Cor e. In a
round-bottom flask, 0.62 g (0.63 mmol) of dipentaerythritol
hexakis(bromoacetate) was dissolved in 10 mL of acetone. To
the clear solution was added tertiary amine (5.1 mmol) in one
portion, and the solution was refluxed for 3 h. During the
reaction, a white precipitate formed gradually and was filtered,
washed several times with acetone, and recrystallized from
CH3CN/acetone to give a white solid.
Dip en t a er yt h r it ol h exa k is(d im et h ylh exyla m m on io-
a ceta te), h exa br om id e: yield 52%; mp ) 161-162 °C; 1H
NMR (CDCl3) δ 5.41 (s, 12H), 4.30 (s, 12H), 3.81 (t, 12H, J )
8.2 Hz), 3.64 (s, 4H), 3.49 (s, 36H), 1.77 (m, 12H), 1.39-1.27
(m, 36H), 0.84 (t, 18H, J ) 7.0 Hz); 13C NMR (CDCl3) δ 165.0,
70.7, 66.0, 65.7, 62.6, 50.8, 42.4, 31.4, 26.0, 22.9, 22.6, 14.1.
Anal. Calcd for C70H142Br6N6O13: C, 47.90; H, 8.15; N, 4.79;
Br, 27.31. Found: C, 47.69; H, 8.16; N, 4.73, Br, 27.53.
Dip en t a er yt h r it ol h exa k is(d im et h yloct yla m m on io-
a ceta te), h exa br om id e: yield 60%; mp ) 163-164 °C; 1H
NMR (CDCl3) δ 5.38 (s, 12H), 4.25 (s, 12H), 3.77 (t, 12H, J )
8.2 Hz), 3.60 (s, 4H), 3.46 (s, 36H), 1.72 (m, 12H), 1.36-1.18
(m, 60H), 0.81 (t, 18H, J ) 6.8 Hz); 13C NMR (CDCl3) δ 164.9,
70.3, 65.8, 65.5, 62.4, 50.6, 42.3, 31.7, 29.2, 29.1, 26.3, 22.8,
22.6, 14.1. Anal. Calcd for C82H166Br6N6O13: C, 51.20; H, 8.70;
N, 4.37; Br, 24.92. Found: C, 51.12; H, 8.75; N, 4.38; Br, 25.04.
Dip en ta er yth r itol h exa k is(d im eth yld od ecyla m m on io-
a ceta te); h exa br om id e: yield 62%; mp ) 167 °C dec; 1H
NMR (CDCl3) δ 5.43 (s, 12H), 4.28 (s, 12H), 3.80 (t, 12H, J )
8.2 Hz), 3.62 (s, 4H), 3.48 (s, 36H), 1.75 (m, 12H), 1.39-1.21
(m, 108H), 0.85 (t, 18H, J ) 6.8 Hz); 13C NMR (CDCl3) δ 164.9,
70.5, 66.0, 65.7, 62.6, 50.7, 42.4, 32.1, 29.8, 29.7, 29.6, 29.6,
1
solid: mp ) 316-317 °C (lit.12 mp ) 317-320 °C); H NMR
(DMSO-d6) δ 4.58 (s, 4H), 1.36 (s, 12H); 13C NMR (DMSO-d6)
δ 68.9, 51.5. Anal. Calcd for C10H16O4: C, 59.98; H, 8.05.
Found: C, 59.85; H, 8.03.
1,3,5,7-Tetr akis(br om oacetoxy)adam an tan e. To a stirred
suspension of 1,3,5,7-tetrahydroxyadamantane (0.59 g, 2.95
mmol) in 0.96 mL (11.8 mmol) of pyridine and 6 mL of CH3-
CN was added a catalytic amount (10 mg) of 4-(dimethylami-
no)pyridine (DMAP). BrCH2COBr (2.06 mL, 23.6 mmol) was
then added dropwise at 0 °C over 1 h. The ice bath was
removed, and the stirring was continued at 50 °C for 5 h, after
which time the solution became transparent and homogeneous.
HCl (6 M, 3 mL) was then added to the reaction mixture
together with 5 mL of CH2Cl2. The aqueous layer was extracted
with CH2Cl2 (3 × 6 mL), and the combined organic layers were
concentrated to give a light-brown solid residue. The solid was
dissolved in 5 mL of CH2Cl2 and washed with brine and with
5 mL of aqueous 0.5 M Na2CO3, after which the organic layer
turned brown-black. This was concentrated and purified by
filtering through a silica gel column. Elution with CH2Cl2 gave
the desired product as a yellow-white solid, which was
recrystallized from CH3CN/MeOH/H2O yielding 1.45 g (82%)
of white needles: mp ) 170.5-171 °C; 1H NMR (CDCl3) δ 3.76
(s, 8H), 2.58 (s, 12H); 13C NMR (CDCl3) δ 166.0, 79.4, 42.8,
26.6. Anal. Calcd for C18H20Br4O8: C, 31.61; H, 2.95; Br, 46.73.
Found: C, 31.89; H, 2.95; Br, 46.46.
Gen er a l P r oced u r e for th e P r ep a r a tion of th e Gem in i
Su r fa cta n ts Ba sed on th e P en ta er yth r itol Cor e. In a
round-bottom flask, 0.54 g (0.87 mmol) of pentaerythritol
tetrakis(bromoacetate) was dissolved in 10 mL of acetone
(CHCl3 was used as a solvent for the preparation of the C8-
chained surfactant). To the clear solution was added tertiary
amine (5.2 mmol) in one portion, and the solution was refluxed
for 3 h. During the reaction, a white precipitate formed
gradually and was filtered and washed several times with
acetone to give a white solid.
29.5, 29.5, 26.5, 23.0, 22.9, 14.4. Anal. Calcd for C106H214
-
Br6N6O13: C, 56.33; H, 9.54; N, 3.72; Br, 21.21. Found: C,
56.49; H, 9.39; N, 3.65; Br, 21.30.
Gen er a l P r oced u r e for th e P r ep a r a tion of th e Gem in i
Su r fa cta n ts Ba sed on th e Ad a m a n ta n e Cor e. In a round-
bottom flask, 0.52 g (0.76 mmol) of 1,3,5,7-tetrakis(bromoac-
etoxy)adamantane was dissolved in 10 mL of acetone. To the
clear solution was added tertiary amine (4.54 mmol) in one
portion, and the solution was vigorously stirred for 3 h. During
the reaction, a white precipitate formed gradually and was
filtered and washed several times with acetone to give a white
solid.
1,3,5,7-T e t r a k is (d im e t h y lh e x y la m m o n io a c e t o x y )-
a d a m a n ta n e, Tetr a br om id e. Recrystallized from CH3CN:
yield 78%; mp ) 199 °C dec; 1H NMR (DMSO-d6) δ 4.50 (s,
8H), 3.48 (t, 8H, J ) 8.0 Hz), 3.21 (s, 24H), 2.59 (s, 12H), 1.67
(m, 8H), 1.31-1.25 (m, 24H), 0.87 (t, 12H, J ) 6.8 Hz); 13C
NMR (DMSO-d6) δ 163.8, 79.5, 64.7, 60.8, 51.0, 42.1, 30.6, 25.3,
21.8, 21.8, 13.8. Anal. Calcd for C50H96Br4N4O8: C, 50.01; H,
8.06; N, 4.67; Br, 26.61. Found: C, 49.78; H, 8.06; N, 4.63; Br,
26.49.
P en t a er yt h r it ol
t et r a k is(d im et h ylh exyla m m on io-
a ceta te), tetr a br om id e: yield 70%; mp ) 152-153 °C; 1H
NMR (CDCl3) δ 5.32 (s, 8H), 4.34 (s, 8H), 3.75 (t, 8H, J ) 8.2
Hz), 3.48 (s, 24H), 1.76 (m, 8H), 1.35-1.25 (m, 24H), 0.84 (t,
12H, J ) 6.8 Hz); 13C NMR (CDCl3) δ 164.7, 66.3, 64.9, 62.4,
51.0, 41.2, 31.3, 25.9, 22.8, 22.5, 14.0. Anal. Calcd for C45H92
-
Br4N4O8: C, 47.54; H, 8.16; N, 4.93; Br, 28.11. Found: C, 47.42;
H, 8.04; N, 4.92; Br, 28.26.
(11) Sollott, G.; Gilbert, E. J . Org. Chem. 1980, 45, 5405.
(12) Stetter, H.; Krause, M. Liebigs Ann. Chem. 1968, 717, 60.

Synthesis and Properties of Multiarmed Geminis
J . Org. Chem., Vol. 64, No. 24, 1999 8921
1,3,5,7-T e t r a k is (d im e t h y lo c t y la m m o n io a c e t o x y )-
Br4N4O8: C, 55.62; H, 9.05; N, 3.93; Br, 22.42. Found: C, 55.35;
H, 9.16; N, 3.97; Br, 22.40.
a d a m a n ta n e, Tetr a br om id e. Recrystallized from CH3CN/
1
acetone: yield 86%; mp ) 201 °C dec; H NMR (DMSO-d6) δ
1,3,5,7-Tet r a k is(d im et h yld od ecyla m m on ioa cet oxy)-
a d a m a n ta n e, Tetr a br om id e. Recrystallized from CH3CN:
yield 87%; mp ) 193 °C dec; 1H NMR (DMSO-d6) δ 4.50 (s,
8H), 3.48 (t, 8H, J ) 8.2 Hz), 3.21 (s, 24H), 2.59 (s, 12H), 1.68
(m, 8H), 1.32-1.21 (m, 72H), 0.85 (t, 12H, J ) 6.8 Hz); 13C
NMR (DMSO-d6) δ 163.8, 79.5, 64.8, 60.8, 50.9, 42.1, 31.3, 29.0,
29.0, 28.9, 28.8, 28.7, 28.5, 25.7, 22.1, 21.8, 14.0. Anal. Calcd
for C74H144Br4N4O8: C, 57.81; H, 9.44; N, 3.64; Br, 20.79.
Found: C, 57.59; H, 9.49; N, 3.49; Br, 20.60.
4.49 (s, 8H), 3.47 (t, 8H, J ) 8.2 Hz), 3.21 (s, 24H), 2.59 (s,
12H), 1.68 (m, 8H), 1.33-1.23 (m, 40H), 0.87 (t, 12H, J ) 6.6
Hz); 13C NMR (DMSO-d6) δ 163.8, 79.5, 64.8, 60.8, 50.9, 42.1,
31.2, 28.4, 28.4, 25.7, 22.1, 21.8, 13.8. Anal. Calcd for C58H112
-
Br4N4O8: C, 53.05; H, 8.60; N, 4.27; Br, 24.34. Found: C, 52.89;
H, 8.63; N, 4.19; Br, 24.20.
1,3,5,7-T e t r a k is (d im e t h y ld e c y la m m o n io a c e t o x y )-
a d a m a n ta n e, Tetr a br om id e. Recrystallized from CH3CN:
yield 90%; mp ) 193 °C dec; 1H NMR (DMSO-d6) δ 4.52 (s,
8H), 3.49 (t, 8H, J ) 8.2 Hz), 3.22 (s, 24H), 2.59 (s, 12H), 1.67
(m, 8H), 1.32-1.21 (m, 56H), 0.85 (t, 12H, J ) 6.8 Hz); 13C
NMR (DMSO-d6) δ 163.8, 79.5, 64.8, 60.8, 50.9, 42.1, 31.3, 28.9,
Ack n ow led gm en t. This work was supported by the
Army Research Office.
28.8, 28.7, 28.5, 25.7, 22.1, 21.8, 14.0. Anal. Calcd for C66H128
-
J O9912350