The effect of size on the rate of an aminolysis reaction using a series of amine terminated PAMAM dendrimers

Article abstract of DOI:10.1016/S0040-4039(02)00113-2

When compared to an equivalent amine concentration of ethylenediamine, the initial rate of a simple aminolysis reaction in water (at pH 8.5) was found to be greatly enhanced by the use of amine terminated PAMAM dendrimers of generations 1-5 (with 4, 8, 16

Full text of DOI:10.1016/S0040-4039(02)00113-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 2431–2433
The effect of size on the rate of an aminolysis reaction using a
series of amine terminated PAMAM dendrimers
John L. Burnett, Amy S. H. King, Ian K. Martin and Lance J. Twyman*
Chemistry Department, Dainton Building, University of Sheffield, Sheffield S3 7HF, UK
Received 6 November 2001; revised 3 January 2002; accepted 14 January 2002
Abstract—When compared to an equivalent amine concentration of ethylenediamine, the initial rate of a simple aminolysis
reaction in water (at pH 8.5) was found to be greatly enhanced by the use of amine terminated PAMAM dendrimers of
generations 1–5 (with 4, 8, 16, 32 and 64 terminal amines, respectively). The fourth generation dendrimer gives the maximum rate
enhancement, even when compared to the larger 64 amine terminated dendrimer. The observed increase in aminolysis rates is due
to hydrophobic binding of the substrate within the outer region of the larger dendrimers, which results in the substrate being held
in close proximity to the reactive amine groups on the surface. Another contributory factor is transition state stabilisation by the
internal amide groups within the dendrimers. The trend in reactivities of dendrimers is discussed and provides strong evidence that
both these effects are responsible for the enhanced rates of aminolysis. © 2002 Published by Elsevier Science Ltd.
Dendrimers are monodisperse, highly symmetrical
macromolecules with a tree-like complexity and a high
density of terminal groups. The surface functionality
dendrimers with polar surface groups as solubilising
agents, which encapsulate hydrophobic guests in their
interiors. Therefore, it is feasible that these two fea-
1
8,9
can be tailored, for example, adapting the terminal
groups to make them more or less hydrophilic can
change their solubility properties. When these terminal
groups are either charged or polar the resulting den-
drimers have similar structural and solubility properties
to micelles. There are numerous reports in the literature
that micelles can be used to catalyse certain reactions
tures can be combined and dendrimers with polar sur-
face groups and an accessible hydrophobic interior can
be used to catalyse or accelerate various reactions.
1
0
Previously, the feasibility of this proposal was demon-
11
strated within our group, when it was shown that the
initial rate of an aminolysis reaction can be greatly
enhanced by using a PAMAM dendrimer with 64 ter-
minal amine groups. This initial rate was compared to
that obtained using 64 equivalents of the simple,
2
3
including hydrolysis, nucleophilic addition, epoxida-
4
5
6
tion, halogenation, and pericyclic reactions. How-
ever, there are some problems associated with their use,
for example, micelles form under equilibrium condi-
tions (so they are constantly collapsing and reforming)
and they can only form above the critical micelle con-
centration (CMC). Dendrimers, however, have the
advantage of being present at all concentrations and
over a wide range of temperature, i.e. they can be
considered as static covalent micelles and this has
mononuclear
amine,
N-acetyl
ethylenediamine
(
NAEDA), which resembles the outer domain of the
dendrimer and therefore acts as an isolated, reactive
unit independent of a dendrimer backbone. A 22-fold
increase in rate was observed with the dendrimer 5
compared to N-acetyl ethylenediamine. It was postu-
lated that there were two factors responsible for this
rate enhancement. Firstly, the dendrimer is acting as a
static covalent micelle, and is solubilising the hydropho-
bic p-nitrophenyl acetate within the outer hydrophobic
region of the dendrimer. Once bound, the p-nitrophenyl
acetate group is held in close proximity to the den-
drimers outer reactive amine groups and this increase in
the effective molarity of the reactive species contributes
significantly to the observed rate enhancement. A sec-
ond factor involves the dendrimers internal amide
groups. It was felt that these may be able to stabilise
the transition state as it forms during the aminolysis
7
important implications for catalysis. Ford has demon-
strated the use of ‘micelle-like’ dendrimers with charged
surface groups for the catalysis of decarboxylation and
hydrolysis reactions, and the enhanced rate here is due
solely to the highly charged nature of the dendrimer
surface, which results in concentration of the hydro-
phobic substrate molecules at the surface. Recent
reports have also described the use of water soluble
*
Corresponding author. Fax: +44-(0)114-273-8673; e-mail:
l.j.twyman@sheffield.ac.uk
0
040-4039/02/$ - see front matter © 2002 Published by Elsevier Science Ltd.
PII: S0040-4039(02)00113-2

2
432
J. L. Burnett et al. / Tetrahedron Letters 43 (2002) 2431–2433
reaction. Such stabilisation effects have been reported
for aminolysis reactions involving other branched multi
amines. In order to gain more information about the
importance of these two factors, we decided to investi-
gate a series of dendrimers from generations 0–5 and
determine at what stage significant rate enhancement
occurs.
tate in acetonitrile was added to the dendrimer/EDA in
aqueous buffer and detection of the product 8 began
immediately. Formation of the product 8 was followed
by measurement of the UV absorbance at 410 nm at
regular time intervals over the course of the reaction
(until saturation was reached). In addition to UV
absorbance measurements, the reaction of the den-
drimer with 64 terminal amine groups (5) was moni-
tored by GC mass spectrometry, which revealed a very
small, insignificant peak for the hydrolysis product,
acetic acid (as expected due to the low rate of back-
ground hydrolysis measured). This confirms that the
reaction occurring is aminolysis (Scheme 1), rather than
a simple hydrolysis catalysed by the presence of the
amine dendrimers.
1
2
The selected aminolysis process studied is shown in
Scheme 1, and involves the reaction between an amine
(
dendrimer or ethylenediamine) and p-nitrophenyl ace-
tate. This is a particularly suitable reaction for investi-
gation as the product 8 is a good leaving group, which
is yellow in colour and can be monitored by measure-
ment of its UV absorbance (umax at 410 nm). Once
again, N-acetyl ethylenediamine was chosen as the con-
trol/test reaction. All reactions were performed in
buffered water at pH 8.5 (Tris buffer 0.1 M) and the
acylated amine is produced as well as the p-nitropheno-
late 8. A control experiment was performed to measure
the background rate of hydrolysis of p-nitrophenyl
acetate in the aqueous buffer. The initial rate for the
background hydrolysis was found to be extremely slow;
nevertheless, all subsequent aminolysis rates were
adjusted to take account of this. A series of experiments
The rate profiles for the reactions clearly demonstrated
that the order of reactivity is EDA=NAEDA<G1<
G2<G3<G5<G4. Using the linear regions of the graphs
obtained during the first 3 min of each reaction (Fig. 1),
initial reaction rates were ascertained and the trend of
dendrimer reactivities as a function of generation num-
14
ber is shown in Table 1. This demonstrates that there
is a rapid increase in the initial aminolysis rate from G0
(EDA) up to generation 4 (which has a 28-fold rate
enhancement relative to the control), then a decrease in
rate is observed for generation 5 with 64 terminal amine
groups (although it is still 21 times more reactive than
the control).
1
3
was carried out with EDA (generation 0) and den-
drimer generations 1–5 (the starting concentration of
amine groups was the same for each experiment). A
solution of known concentration of p-nitrophenyl ace-
O
O-
O
O
Me
aq pH 8.5
R
PAMAM
Generations
NH2
+
N
H
Me
O N
N
H
+
O
n
2
1
- 5
O2N
8
or
7
R = PAMAM
H2N
1
2
3
4
5
generation 1, n = 4
generation 2, n = 8
generation 3, n = 16
generation 4, n = 32
generation 5, n = 64
NH2
or
6
H2N
Scheme 1. Aminolysis of p-nitrophenyl acetate.
Figure 1. Linear regions of the curves of concentration versus time for EDA and 1–5 during the aminolysis reaction.

J. L. Burnett et al. / Tetrahedron Letters 43 (2002) 2431–2433
2433
Table 1. Initial rates of aminolysis for EDA and 1–5 and initial rates relative to EDA
Generation
Amines per molecule
Concentration (mM)a
Initial rate (M s−1
)
Relative rate to EDA
−
−
−
−
−
−
10
9
0
1
2
3
4
5
2
4
8
16
32
64
3.2
1.6
0.8
0.4
0.2
0.1
4.03×10
2.40×10
5.20×10
9.02×10
1.12×10
8.63×10
1
5.95
12.89
22.35
27.56
21.4
9
9
8
9
a
Relative amine concentration is 6.4 mM in each case (and p-nitrophenyl acetate concentration is 0.014 mM).
As expected, the results confirm that there is a hydro-
phobic effect operating here, i.e. the larger dendrimers
molecules to the hydrophobic environment within the
dendrimer.
(
>generation 2) are acting as static covalent micelles
and solubilising the p-nitrophenyl acetate within the
inner hydrophobic regions of the dendrimers. This
increase in effective molarity of the reagents therefore
contributes significantly to the observed rate enhance-
ments in the larger dendrimers. As postulated previ-
ously, the rate enhancements can also be attributed to a
second factor involving the dendrimers internal amide
groups, which are able to stabilise the tetrahedral tran-
sition state formed between the amine and the p-nitro-
phenyl acetate. The presence of this transition state
stabilisation effect accounts for the fact that the small
dendrimers (generations 1 and 2) have faster aminolysis
rates than the simple amine, EDA (G0). For example,
the first generation dendrimer (four amines) is six times
more reactive than EDA, despite the fact that it is still
effectively a small molecule that does not provide a
hydrophobic environment for substrate molecules.
Thus, it is clear that there is no hydrophobic effect
operating and the substantial increase in rate is due
solely to the ability of the first generation dendrimer to
form a tetrahedral intramolecular transition state,
which lowers the energy of activation (DG) for the
aminolysis reaction and results in a faster rate. As the
dendrimers increase in size, the increase in initial rates
continues (up to generation 4). Therefore, we conclude
that this increase in rate is due to transition state
stabilisation and hydrophobic binding.
References
1. Zeng, X.; Cheng, Y. J. Dispers. Sci. Technol. 2000, 21,
449.
2. (a) Cheng, S.; Zeng, X. J. Dispers. Sci. Technol. 2000, 21,
655; (b) Cordes, E. H. Reaction Kinetics in Micelles;
Plenum Press: New York, 1973.
3. (a) Bunton, C. A.; Bacaloglu, R. J. Colloid Interf. Sci.
1987, 115, 288; (b) Baumucker, J.; Calzadilla, M.; Cen-
teno, M.; Lehrmann, G.; Urdanela, M.; Lindquist, P.;
Dunham, D.; Price, M.; Sears, D.; Cordes, E. H. J. Am.
Chem. Soc. 1972, 94, 8164; (c) Swiswanto, C.; Battal, T.;
Schuss, O. F.; Rathman, J. F. Langmuir 1997, 13, 6047.
4. Monti, D.; Tagliatesta, P.; Mancini, G.; Boschi, T.
Angew. Chem., Int. Ed. Engl. 1998, 37, 1131.
5. Jaeger, D. A.; Robertson, R. E. J. Org. Chem. 1977, 42,
3298.
6. (a) Van Mersbergen, D.; Wijnen, J. W.; Engberts, J. B. F.
N. J. Org. Chem. 1998, 63, 8801; (b) Jaeger, D. A.; Wang,
D. Tetrahedron Lett. 1992, 33, 6415; (c) Newkome, G. R.;
Moorfield, C. N.; V o¨ gtle, F. Dendritic Molecules: Con-
cepts, Synthesis and Perspectives; VCH: Weinheim, 1996.
7. Lee, J. J.; Ford, W. T. Macromolecules 1994, 27, 4632.
8. (a) Twyman, L. J.; Beezer, A. E.; Esfand, R.; Hardy, M.
J.; Mitchell, J. C. Tetrahedron Lett. 1999, 40, 1743; (b)
Esfand, R.; Beezer, A. E.; Mitchell, J. C.; Twyman, L. J.
Pharm. Sci. 1996, 2, 157.
This observed trend in reactivity was not surprising,
considering the properties of the PAMAM dendrimers.
It has been reported in the literature that there is a
transition from an extended to a globular closed shell
conformation for related amine terminated den-
9. Hawker, C. J.; Wooley, K. L.; Fr e´ chet, J. M. J. J. Chem.
Soc., Perkin Trans. 1 1993, 1287.
10. Piotti, M. E.; Rivera, F.; Bond, R.; Hawker, C. J.;
Fr e´ chet, J. M. J. J. Am. Chem. Soc. 1999, 121, 9471.
11. Martin, I. K.; Twyman, L. J. Tetrahedron Lett. 2001, 42,
1123.
12. Evans, D. J.; Kanagasooriam, A.; Williams, A. J. Mol.
Catal. 1993, 85, 21.
13. An acetonitrile solution of p-nitrophenyl acetate (0.14 M,
0.1 ml) was added to a solution of amine (6.4 mM w.r.t.
amine groups) in buffered water at pH 8.5 (0.1 M TRIS
buffer, 1.9 ml) in a quartz UV cuvette (with stirring, at
room temperature) and the UV absorbance at 410 nm
was measured every 30 s over a period of 20 min.
14. The mechanism is complicated and we were unable to fit
the data to simple first- or second-order kinetics therefore
initial rates were used.
1
5
drimers. Similarly, we have observed that there is a
change in behaviour, i.e. a break in the pattern regard-
ing rate enhancement between generation 4 (32 amines)
and 5 (64 amines). The fourth generation dendrimer
still has a relatively open and flexible structure, which
allows substrate molecules to enter and reside in the
large hydrophobic interior (and facilitates the forma-
tion of the T transition state), so a maximum in the
d
rate enhancement for this dendrimer is observed, i.e.
the conditions are optimum for aminolysis rate
enhancement. The initial rate with the 64 amine den-
drimer is lower than for generation 4 (and similar to
that observed for generation 3), which is presumably
due to the fact that the surface of the dendrimer is
highly crowded preventing access of the substrate
15. Jansen, J. F. J. A.; Meijer, E. W. J. Am. Chem. Soc. 1995,
117, 4417.