National Sun Yat-sen University Institutional Repository:Item 987654321/27582

行政院國家科學委員會專題研究計畫 成果報告

單分子螢光技術應用於奈米結構之型態與性質之研究

中 華 民 國 95 年 6 月 16 日

一、中文摘要 本年度計劃的目標是完成單分子光譜實驗設備架設，以及進行初步實驗。在實驗成 果上，我們主要完成了(1)進行單一量子點的光譜及時間解析光譜之測量，以及這些光 物理性質之分佈 (2)進行 cyanine 染料之單分子光譜測量，並且比較不同偏極光分量之 比值，發現在光激發過程中發生螢光偏極方向改變之現象，並推斷這是由於光激發導致 構型變化(photo-isomerization)。特別的是，我們發現其中有需時達到~10ms 時間長度 的構型變化。 關鍵詞：量子點、單分子螢光、光激發導致構型變化 Abstract

The goal for this year’s project is to accomplish the setup for single molecule

fluorescence measurements, and perform initial experiments. We achieve it, and we studied semiconductor quantum dots, and cyanine dye molecules. The results include (1) study

individual quantum dot’s fluorescence spectrum and the fluorescence lifetime measurements,

and the distribution of the properties. (2) Single molecule investigating cyanine dye molecules. Through a polarization beamsplitter, we found that the fluorescence anisotropy has stepwise changed during the measurement, and we conclude it from photo-induced isomerization. In particular, some isomerized process takes ~ 10 ms, which is much longer than expected.

二、報告內容

We accomplish the setup for single

molecule experiments. The basic

configuration is shown below, which is already shown in our last year’s report. Nevertheless, there is minor change to allow us a better working system. The system provides us simultaneously measure the fluorescence intermittence, fluorescence anisotropy, decay lifetime, from a single molecule (single object).

In addition, we accomplish some

experiments. Two major achievements

include (1) study the size distribution of the CdSe/ZnS quantum dots, and measure the corresponding fluorescence properties. (2) Through a polarization beamsplitter, we study the transition dipole moment orientation.

Many single molecules exhibit consistent transition dipole moment orientation, but some show stepwise orientation change. The results imply a photo-induced isomerization occurrence. In particular, we found that some isomerization process takes ~ 10ms, which is much longer than expected.

In the first part, we study the core-shell

structure semiconductor quantum dots:

CdSe/ZnS. The quantum dots sample was purchased from Evident Technologies, which is a toluene solution with the size of 5.2 nm

in diameter. Basic properties can be found in the web site.[1] Figure 1 shows the typical fluorescence spectrum for a single quantum dot recording of 10 second. The PL spectrum has a peak at 612.4 nm, and spectral width of 17.8 nm. With collecting of 65 quantum dots results, we got the distribution, which are shown in Figure 2 and 3. With one exception, the fluorescence peaks exhibit a good Gaussian distribution of the center at 615.7 nm and a distribution width of 13.8 nm. In addition, the spectral width also exhibits a Gaussian

550 600 650 700 750 800 0 30 60 90 120 150 In te ns it y Wavelength (nm) Peak 612.4 nm Width 17.8 nm Fig. 1 16 18 20 22 24 Q D Fitted 615.7 nm ? 13.8 nm 16 18 20 22 24 D 19.7 nm6.9 nm

distribution with average width 19.7 nm.

The result is consistent with the ensemble average measurement done in

the 10-8M toluene solution. Ensemble

average measurement got a fluorescence spectrum with center at 616 nm, and a width of 25 nm. This is consistent with the quantum dots distribution of peak central at 615.7 nm and the width of

nm

.

nm

.

nm

.

8

19

7

24

0

13





In addition, we have performed the Transmission Electron Microscopy (TEM) measurements. According to the aspect ration of the long axis length and the short axis length, the quantum dots were classified into 5 different classes. Most of

the quantum dots belong to the

quasi-spherical shape with the aspect ration within 1.1 to 1.5, which is about 76% in our sample. More than 97% of the quantum dots have the aspect ratio between 1.0 and 2.0.

Not only the shape classification, we also analysis the distribution of the long axis and short axis lengths. The result indicates both are able to have good fitting with the Gaussian distributions. The long axis has a distribution of peak at 5.4 nm and width of 1.3 nm, and the short axis has a distribution of peak at 4.6 nm and width of 1.0 nm.

In order to have better understanding of the size-spectrum relationship, we convert

the fluorescence spectrum into the size distribution by using previously built relationship, and compared with the distribution from TEM measurement.[2] The result indicates that the size derived from the fluorescence

peak is much narrower than the actual size measured by TEM. Further modification by consideration of the non-spherical shape, we fit the size distribution with the geometric mean, which is defined as



long

axis

length





short

axis

length



, and we get a more consistent fitting result. Thus we conclude that in a non-spherical shape quantum dot, it is not

2 3 4 5 6 7 8 9 10 0 50 100 150 200 250 300 350 400 個 數 粒子直徑（nm） 長軸：peak=5.4nm 半高寬=1.3nm 短軸：peak=4.6nm 半高寬=1.0nm 2 3 4 5 6 7 8 9 10 0 50 100 150 200 250 300 350 400 450 由螢光峰值所得粒徑分布 TEM結果 長軸分布 短軸分布 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 0.0 0.2 0.4 0.6 0.8 1.0 N or m al iz ed D is tr ib ut io n Mean Diameter 螢光轉換粒徑分佈 peak=4.8nm 半高寬=0.7nm 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 0.0 0.2 0.4 0.6 0.8 1.0 量子點粒徑分佈_peak=4.8nm 半高寬=0.9nm 三角形 球形（1<s<1.1） 類球形（1.1=<s<1.5） 橢球形（1.5=<s<2） 橢球形（s>=2） 0 200 400 600 800 1000 1200 個數

only the long axis length to influence the fluorescence spectrum, but also the short axis length. Geometric mean can be a good estimation for the mean diameter for the non-spherical shape nano-particles.

The second result is the observation of fluorescence anisotropy of a cyanine dye molecule. The molecule we investigated is DiI, a typical dye molecule used in single molecule investigation, which belongs to the cyanine dye molecules class. The generic cyanine dyes consist two nitrogen centers, one of which is positively charged and is linked by a conjugated chain of an odd number of carbon atoms to the other nitrogen. Cyanine dye molecules have been long interesting for photo-sensitizers in photography, optical storage media, nonlinear optical materials, laser dyes, and bio-labeling dye molecules. Generally, the dyes have an all-trans geometry in their stable form. Occasionally these dyes undergo photo-isomerization, which is one of the most important processes in photo-chemical and photo-biological processes.[3]

DiI is interested in the bio-luminescence labeling. In particular, due to its high photostability, DiI is among one of the molecules extensively study by single molecule

fluorescence. Typically, a DiI molecule can undergo 108 excitation cycling before it is

photobleached. However, there is few reports on the photo-isomerization in the molecules. Photo-isomerization is one process that influences photostability. When a molecule is easily photo-isomerized, usually it represents less photo-stable. However, many reports concern the photoisomerization of cyanine dye molecules, it is less likely that DiI is highly stable and free from the photo-isomerization process. In addition, DiI and analogous molecules exhibit less photo-stability in the solution phase, which imply the possibility to have higher photo-isomerization probability. When the molecule embedding in the polymer matrix, it restricts the flip motion associated with the photo-isomerization process, which can be one reason to have high photo-stability and less photo-isomerization processes.

Usually the cis form is less stable, and has higher energy level. It was believed that when the molecule undergoes photo-isomerization, it changes the energy levels, and shifts the absorption and fluorescence spectra. There is also associated change of the transition dipole strength, and thus the fluorescence quantum yield. In short, when a molecule undergoes photo-isomerization process, the fluorescence was believed to be quenched. However, recently investigation found that an analogous molecule ‘CY5” can have several conformers with similar bandgap, and the oscillator strength.[4] This imply that when DiI undergoes photo-isomerization process, it might not turn “on” state into “off” state (otherwise it is not possible to be distinguished from the photo-bleaching process). Moreover, the associated conformation change may reflect on the polarization anisotropy.

We perform the single molecule fluorescence studies on DiI molecules embedding in PMMA matrix. The fluorescence was through a polarization beamsplitter, and then sent into two avalanche photodiodes for the polarization anisotropic information.

typical result is shown in the right figure Under strong laser illumination, the molecule survives around 133 seconds, and it exhibits clear on-off blinking behavior, which is extensively studied.[5] The fluorescence anisotropy, defined as

II II

I

I

I

I





trajectory of the monitored molecule orientation. Since the molecule is embedded in the polymer matrix, it is expected to have a fix molecule orientation, and hence a constant fluorescence anisotropy during the measurement. However, some molecules exhibit stepwise change of the anisotropy, which reflects the sudden change of molecule orientation.

The next figure shows a typical result for the observation of polarization change. The molecule undergoes ~ 30 second of laser excitation. During the measurement, the polarization change while exhibits negligible intensity change. In particular, the anisotropy jumps 0 and -0.7, which is corresponding to stepwise reversible processes.

When we further look into a more detail, the result provides strong support for a single molecule photoisomerization processes. Shown in the follow figure, The intensity from two detectors are labeled with red and blue color. Clear simultaneous on-off blinking behavior indicates both signals from the same single molecule. When we integrate the decay dynamics in a period of the same anisotropy, we can further analysis the average decay lifetime in the period. The decay lifetimes are able to be fitted by the single decay component, and the fitted lifetimes are listed in the figure. It results a consistent relation between the decay lifetime and anisotropy. In particular, we found that at time ~18.7 second, there is no signature indicating off-states, the molecule continuing emits photons. But the polarization anisotropy continues change, which lasts ~ 10 ms. Other polarization change processes occur at the off-states, or relates to the off-states, which we are not able to conclude the photo-isomerization processes happened in the singlet excited states or the triplet states. However, the observation at 18.7 second clearly identify the process occurred in the singlet state, In addition, it takes ~ 10 ms, which is more longer than our expectation. The much longer time for the process might be due to the highly viscous environment, which hinders the isomerization process. More studies on the power dependence are still undergoing to have a concluding result.

Reference:

1. Evident Technologies web site:http://www.evidenttech.com/

2. Yu W.W., Qu L., Guo W., Peng X., Chem. Mater, 15, 2854-2860 (2003)

3. Mishra A., Behera R.K., Behera P.K., Mishra B.K., Behera G.B., Chem. Rev., 100, 1973-2011 (2000)

4. Vallee R.A.L., Marsal P., Braeken E., Habuchi S., Schryver F.C.D., Van der Auweraer M., Beljonne D., Hofkens J., J. Am. Chem. Soc. 127 12011-12020 (2005) 5. for example, Hurner C.G., Renn A., Renge I., Wild U.P., J. Chem. Phys. 115,