Supporting Information - rsc.org fileSupporting Information . Temperature Dependent Excited State Relaxation of a Red Emitting DNA-Templated Silver Nanocluster . Cecilia Cerretani,
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Supporting Information
Temperature Dependent Excited State Relaxation of a Red
Emitting DNA-Templated Silver Nanocluster. Cecilia Cerretani,a Miguel R. Carro-Temboury,a Stefan Krause,a Sidsel Ammitzbøll Bogha and Tom
Voscha,*
a Nanoscience Center and Department of Chemistry, University of Copenhagen, Universitetsparken
radiation and a halogen lamp for visible and near infrared radiation. Steady-state fluorescence
measurements were done using either a Fluotime300 instrument from PicoQuant with a 561 nm
laser as excitation source, or using a QuantaMaster400 from PTI with a Xenon arc lamp as
excitation source. All fluorescence spectra were corrected for the wavelength dependency of the
detector system.
Time-Correlated Single Photon Counting. Time-resolved fluorescence measurements were
conducted using the Fluotime300 instrument from PicoQuant with a 561 nm laser as excitation
source. All time-resolved data was fitted globally with a tri-exponential reconvolution model
including scatter light contribution and the IRF in Fluofit v.4.6 from PicoQuant. The average decay
time <τ> of a decay curve was calculated as the intensity-weighted average decay time. The
intensity weighted average decay time <τω> was calculated as the average of <τ> over the
emission spectra weighted by the steady state intensity.
Acquisition and analysis of TRES data. Time-resolved emission spectra (TRES) were acquired by
changing the emission monochromator in steps of 5 nm, with an integration time of 60 s per decay
in order to achieve >10000 counts over the emission spectra. All time-resolved data was fitted
globally with a tri-exponential reconvolution model including scattered light contribution and the
IRF in Fluofit v.4.6 from PicoQuant. The obtained TRES was corrected for the detector efficiency
and transformed to wavenumber units by multiplying with the Jacobian1 107/ν2. The TRES was
then interpolated with a spline function using the built-in spaps MATLAB function using a
tolerance of 10-10 (forcing the interpolated curve to go through the data points). The curve was
interpolated using wavenumber steps equivalent to 0.01 nm steps. The emission maxima were
taken as the maxima of the interpolated TRES.
Fitting details for Figure 4.
Figure 4B: χ2 values for the four cycles 5°C : 1.0562 ; 1.0518 ; 1.0485 ; 1.0417
25°C: 1.1584 ; 1.2070 ; 1.1811 ; 1.1737
40°C: 1.2260 ; 1.0505 ; 1.2339 ; 1.1565
Figure 4D: χ2 values for the single decays from 5 °C - 60 °C in steps of 5 °C (see also Figure S8): 1.0092; 1.0512; 0.9562; 1.0857; 1.1296; 0.9220; 0.9819; 1.0339; 1.0494; 1.0215; 1.0390; 1.1515
Figure S1: HPLC Chromatograms of DNA-AgNCs. A) Absorption at 260 nm as a function of elution time. The 260 nm absorption monitors the DNA absorption. B) Absorption at 570 nm as a function of elution time. The 570 nm absorption monitors the AgNC absorption. C) Emission at 630 nm as a function of elution time. The 630 nm monitors the AgNC emission. The flat top is due to detector saturation.
Figure S2: Steady-state 2D excitation versus emission plot of the purified DNA-AgNCs at room temperature. A) In linear intensity scale. B) In logarithmic intensity scale.
Figure S3: Stability of the purified DNA-AgNCs over time. Absorption spectrum right after purification (black line) and the same solution one month later (red line). The sample was stored in the fridge between the measurements.
Figure S4: Decay associated spectra of DNA-AgNCs at 25°C, first cycle, excited at 561 nm. The grey line indicates the zero line. The intensity decays were acquired every 5 nm, from 575 nm to 725 nm. The three lifetime values were globally linked in the fit. The global fit had a reduced χ2 of 1.1584.
Analysis of QY data and derivation of the QY equations.
The measured QY was plotted against the calculated <τω> for the measurements at different
temperatures (data from Cycle 1) and fitted with a linear model of the form (1)
QY(<τω>)=a<τω>-b
where a,b are positive constants (Fig. SQY). The variable <τω> was plotted against the temperature
T and fitted with a linear model of the form (2)
<τω>(T)=n-mT
where n,m are positive constants (Fig. SQY).
This implies that (3)
QY(T)=p-qT
where p=an-b and q=ma are positive constants. In this way we can relate all the constants of the
predictions derived from the linear regressions and use equations (2) and (3) to evaluate the
previous quantities in an extended temperature range. Note that this is a calculated extrapolation
and that parts of the extrapolation might not be physically realistic.
Fig. SQY A) Linear regression from measured data parameters a=0.77±0.05 ns-1 and b=1.2±0.1. B)
Linear regression from measured data used to extract the parameters m=0.0067±0.0005 ns°C-1
and n=2.74±0.01 ns.
Putting together the equations from the model presented in the manuscript, equation (4)
QY=QYS1·QYf
and equation (5)
QYf=kf·<τω>
, we can obtain (6)
QY=QYS1· kf·<τω>
Since we measured experimentally QY for the same values of T as <τω>, we know that in the
measured range the dependency QY(<τω>)= f(<τω>) described by (1) holds, and putting (1) and (6)
together we can deduce that (7)
QYS1·kf=a-b/<τω>
The observed linear behavior seems to predict an extended region, as evidenced from Fig. 4 in the
manuscript and intensity data (not shown). However, if a different functional dependence
QY(<τω>)= f(<τω>) than (1) is observed outside the measured range, the same formalism here
described can be used inserting the new function f(<τω>). Since it is not possible from equation (7)
to know whether kf or QYS1 are responsible for the observed data described by equation (1), we
distinguished three possible scenarios. In all the scenarios we can calculate the non-radiative rate
from (8)
knr=1/<τω>-kf.
kf is assumed to be constant and with the value kf100%= a/(1+b)= 3.5 108 s-1. The value of QYS1 is
then calculated using equation (7) and is fully responsible for the observed QY(<τω>). QYS1
decreases faster than QYf as a function of temperature.
Scenario I
Figure S5:
Scenario I. A) QY versus <τω>. B) QY versus T. C) kf, knr, ktot versus T.
QYS1 is assumed to be constant and with the value QYS1=1. This value was chosen since QY=0.9 at 5
°C and it therefore seems reasonable that a value for QY=1 is achievable. In this case, due to (4)
QYS1≥QY the assumption QYS1=1 seems to be a likely option. The value of kf is then calculated using
equation (5) and the drop of QYf is fully responsible for the observed decrease of QY as a function
of temperature.
Scenario II
Figure S6: Scenario II. A) QY versus <τω>. The red and blue curves overlap. B) QY versus T. The red and blue curves overlap. C) kf, knr, ktot versus T.
Both QYS1 and kf are variable in equation (7). In this case, following the reasoning explained in
scenario II, the option of QYS1 increasing as a function of temperature is discarded (it can’t
physically go above 1). The two other options left are III (a) both kf and QYS1 decrease as a function
of temperature or III (b) kf increases as a function of temperature and QYS1 compensates with a
faster decrease. As an example, we have analyzed the situation III (a) using a linear decrease of the
form QYS1=k-rT where k and r are positive constants. The value of r was chosen to be r=0.002 °C-1
for this simulation, and the constant k needs to be constrained by the previous equations. In
particular it is required that k=1+rTmin , where Tmin=(p-1)/q=-17.1°C is the temperature at QY=1. In
this case, the QYf decreases faster than QYS1 as a function of temperature.
Scenario III
Figure S7: Scenario III. A) QY versus <τω>. B) QY versus T. C) kf, knr, ktot versus T.
Figure S8: Single fluorescence intensity decay curves at λex=561 nm and λem=633 nm for the DNA-AgNCs sample at different temperatures. Y-Axis is in logarithmic scale.
Calculation of sensitivity.
The estimated sensitivities with respect to the highest decay time were estimated using the values ∆<τ> = 2.1 ns - 0.5 ns, ∆T= 70 °C - 10 °C, <τ>max = 2.1 ns , for Rhodamine B;2 and the values ∆<τ>/ ∆T = 0.0071 ns °C-1 and <τ>max=2.62 ns for the DNA-AgNC.
1. J. Mooney and P. Kambhampati, The Journal of Physical Chemistry Letters, 2013, 4, 3316-3318. 2. R. K. P. Benninger, Y. Koç, O. Hofmann, J. Requejo-Isidro, M. A. A. Neil, P. M. W. French and A. J.