Communications
ions) in basic solution can compete with phenolate
groups on MC to bind with Cu2+. When the pH value
reached 11.0, the binding between Cu2+ and MC on
Au surfaces cannot take place, and AuNPs remained
well-dispersed in aqueous solutions. Thus the optimal
pH range for this logic system was from 4.0 to 8.0
(Supporting Information, Figure S11).
To investigate the influence of the length of the
units of EG that tethered spiropyran and alkanethiol,
we synthesized another two spiropyran ligands con-
taining various lengths of the units of EG (Support-
ing Information, Scheme S2 and S3), one having no
EG (C11SP; Scheme 1), and the other containing six
EGs (EG6SP; Scheme 1). We used EG3 as the
stabilizing agents for all three kinds of spiropyran
Figure 3. a) UV/Vis absorption spectra reflecting the cycles of UV/visible light
irradiation (from a to f, 30 s each). b) The sizes (diameter d) of dispersed (A, C,
E) and aggregated (B, D, F) spiropyran-AuNPs measured by DLS.
spectra confirmed these reversible processes; that is, with the
formation of AuNP aggregates, the absorption band at 520 nm
decreased along with the appearance of a new shoulder
absorption band at about 670 nm (Figure 3a, curves of B, D,
F). With the redispersion of aggregates after visible light
irradiation, the absorption band at 520 nm increased along
with the decrease of those at 670 nm (Figure 3a, curves of A,
C, E). The reversible modulations were also supported by the
DLS data (Figure 3b). The average hydrodynamic diameter
of well-dispersed spiropyran-AuNPs is about 24 nm, while
that of their aggregates increased to be approximately
300 nm, congruent with the TEM analyses.
We also used zeta potential measurements to compare the
surface charge on monodispersed spiropyran-AuNPs and
their aggregates at pH 7.0. The zeta potential of well-
dispersed spiropyran-AuNPs was about À30 mV, while that
of their aggregates was approximately 0 mV (Supporting
Information, Figure S9). We reasoned that the increase of
zeta potentials was due to the positively charged ammonium
group on MC isomers, where the phenolate anionic groups
chelated with Cu2+. These results from zeta potentials are
consistent with those from UV/Vis spectroscopy, TEM, and
DLS data. Importantly, we believe that most of the means for
characterization, such as UV/Vis spectra, zeta potential, and
DLS can serve as output signals in the logic gate, by which a
multi-readout logic system could be achieved.
Several factors influence this logic system. First, pH values
of the solution can greatly influence the stability of spiro-
pyran-AuNPs. We prepared aqueous solutions containing
spiropyran-AuNPs with pH values from 2.0 to 12.0. When the
pH values of the solutions decreased below 4.0, the A670/A520
values was much higher owing to the aggregation of AuNPs.
The ether units of EG3 residing on different AuNPs can
protonate, resulting in aggregation.[12] The spiropyran-AuNPs
were well-dispersed in aqueous solutions with pH values
higher than 4.0 (Supporting Information, Figure S10). Next,
we investigated the influence of pH values to this aggregation
by characterizing the absorbance responses to various pH
values. In the range between pH 4.0 and 8.0, the functional-
ized AuNPs, capped by the mixture of EG3SP and EG3 with a
ratio of 1:10, aggregated in tens of seconds after UV light
irradiation and incubation with Cu2+. We note that aggrega-
tion of spiropyran-AuNPs was incomplete with the increase of
pH values from 8.0, because the negative charges (hydroxide
ligands (the ratio was 1:10 for all the three ligands to EG3).
For the MC-AuNPs having no EG (C11mC-AuNPs), almost
no aggregation was observed upon the addition of Cu2+
(10 mm). We reasoned that the stabilizing agents (EG3) on
AuNP surfaces were too long to allow chelation between Cu2+
and MC on different C11mC-AuNPs surfaces. On the other
hand, when we increased the units of EG to six to form
EG6MC-AuNPs, the aggregation was incomplete in the
present of Cu2+ (10 mm). We reasoned that the EG6 tether
was longer than EG3, thus providing more flexibility for MC
groups on EG6MC-AuNP surfaces; some MC groups on
different EG6MC-AuNPs bind with Cu2+ to enhance the
chances of aggregation, while the other MC groups on the
same EG6MC-AuNPs can also have room to bind with Cu2+
owing to the flexible and long tether, thus resulting in reduced
aggregation. Therefore, EG3SP was the best spiropyran
ligand for this logic system (Supporting Information, Fig-
ure S12).
We finally investigated the effect of irradiation time with
UV light. Photochemistry on a solid surface was different
from that in solution,[13] thus the time required for isomer-
ization for SP to MC on AuNP surface was not consistent with
those in solutions. The spiropyran-AuNPs started to aggre-
gate after 10 s of UV light irradiation, and the aggregation
reached its maximum after a persistent irradiation for about
30 s, plateauing thereafter (Supporting Information, Fig-
ure S13), which is slower than that in solution (20 s; Support-
ing Information, Figure S1). This effect is probably due to the
surface quenching and steric constraints of spiropyran on
AuNP surface that decrease the isomerisation efficiency.[13c]
The optimal irradiation time for spiropyran-AuNPs was thus
about 30 s.
In conclusion, we have presented a spiropyran-AuNP-
based, resettable, multi-readout logic system that includes
AND, OR, and INHIBIT logic operations. The distinctive
advantage of this system is that molecular events in aqueous
solution could be translated into a color change of the
solution, which can be monitored by several readouts, such as
UV/Vis spectroscopy, zeta potential, DLS, and even the
naked eye. Importantly, these logic gates were carried out in
aqueous media, potentiating its applications in biological and
biomedical systems.[14] Moreover, the set–reset function of
these logic gates provides memory elements in an all-aqueous
system, ensuring the entire operation is environmentally
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 4103 –4107