provides the chemical structures and acronyms of the conjugated polymers employed in this study (see Experimental section for the full names of polymers). Multicolored CPdots were prepared using the reprecipitation procedure described previously.18
The procedure involves a rapid injection of a hydrophobic polymer in tetrahydrofuran (THF) solution into water, while applying sonication to improve mixing. The sudden decrease in solution hydrophobicity leads to collapse of polymer chains, resulting in the formation of hydrophobic polymer nanoparticles. Since the reprecipitation process involves a competition between aggregation and chain collapse to form nanoparticles, the particle size can be controlled by adjusting the polymer concentration in the precursor solution. To explore the effect of precursor concentration on particle size, we focus our discussion on two polymers that form highly photostable nanoparticles: PPE and PFBT. PPE dots were prepared using different concentrations of the polymer in the precursor solution ranging from 20 ppm to 2000 ppm (see Experimental section for details). As shown in the AFM image of , the PPE nanoparticles exhibit an approximately spherical shape. Analysis of particle height indicates that the nanoparticles produced using the lowest precursor concentration possess small particle size and a relatively narrow size distribution (8±1 nm diameter), while those prepared from more concentrated precursor solutions exhibited larger diameters. PFBT dots that were also prepared using a range of concentrations of the precursor solution exhibit a similar trend — the PFBT dots obtained using a dilute solution exhibit relatively small particle size and size distribution (10±3 nm), while those prepared using higher precursor concentrations exhibit larger size (). The preparation of small particles of both PPE and PFBT polymers are consistent with our previous reports on PDHF, PFPV and MEH-PPV particles.18
The specific particle size and size distribution are also dependent on the polymer species (as well as other conditions that may affect the mixing and aggregation rates). For example, the larger PFBT dots in exhibit a relatively broad size distribution (25±10 nm) and a substantial fraction of small dots. PFPV dots prepared under the same conditions exhibited much larger particle sizes, in the range of 50~70 nm. The size differences are likely due to differences in the molecular weight, polydispersity, and interchain interactions of the different polymers. It was also found that the preparation conditions affect the polymer phase for PFO polymer, with a fraction of β-phase formed under certain conditions.21
Figure 1 (a) Chemical structure of fluorescent conjugated polymer dots; (b) A typical AFM image of small PPE dots and histogram of particle height of the PPE dots prepared with different precursor concentration; (c) A typical AFM image of small PFBT dots and histograms (more ...)
The aqueous CPdot suspensions obtained from various conjugated polymers are stable and clear (not turbid), presenting colors associated with their visible absorption spectra. Under UV lamp illumination (365nm), the nanoparticle dispersions exhibit strong fluorescence with a wide variety of colors, as shown in . The changes in the absorption spectra and fluorescence spectra upon particle formation vary depending on the polymer (). For PFBT, PFPV, and MEHPPV, the absorption spectra are broadened and blue-shifted as compared to those of the polymer in THF solution, which is consistent with an overall decrease in the conjugation length due to bending, torsion, and kinking of the polymer backbone. The nanoparticles also (in most cases) exhibit a slightly red-shifted fluorescence and a long red tail as compared to the polymer in organic solvent. This is attributable to increased interchain interactions due to chain collapse, resulting in a fraction of red-shifted aggregate species.18
The resulting energetic disorder, combined with multiple energy transfer, would result in a net red-shift in the fluorescence spectrum as compared to that of the polymer in THF solution, as has often been observed in thin films.22
For PFO nanoparticles, the additional absorption feature and the red-shifted fluorescence are attributed to the β-phase conformation.21
The relatively complex absorption and fluorescence structures observed for PPE dots may be due to the complex packing and phase behavior of polymer chains, as was previously observed in thin films of similar PPE derivatives.23
However, due to the complex phase behavior of PPE derivatives, additional studies are required to obtain definitive information about the internal structure of the nanoparticles.
Absorption spectra (a), fluorescence spectra (b), photobleaching curves (c), and fluorescence decay traces (d) of various conjugated polymer dots.
CPdots exhibit broad absorption bands ranging from 350 nm to 550 nm (depending on the polymer), a wavelength range that is convenient for fluorescence microscopy and laser excitation. Analysis of the UV-Vis absorption spectra at a known particle concentration indicated that the peak absorption cross section of single particles (~15 nm diameter) were on the order of ~10-13
, roughly ten to hundred times larger than that of CdSe quantum dots in the visible and near-UV range, and roughly three orders of magnitude larger than typical organic fluorescent dyes. Fluorescence quantum yields (ϕF
) ranged from a few percent to as high as 40%, depending on the polymer. The absorption cross section and quantum yield results are summarized in . These results indicate that, to the best of our knowledge, the fluorescence brightness of the CPdot nanoparticles exceeds that of any other nanoparticle of the same size under typical conditions. The size of the particle does not appear to have an appreciable effect on the shape of the absorption and fluorescence spectra — the principal effect of increased particle size is an increase in the absorption cross-section. This property facilitates adjustment of particle size and brightness to meet the demands of a particular application, and is in contrast with colloidal semiconductor quantum dots, which exhibit pronounced variations in band gap due to the quantum size effect. For a given polymer, particle size was also found to affect fluorescence quantum yield. For example, a quantum yield of ~0.10 was determined for small PFPV dots (~10 nm), while larger PFPV particles (~60 nm) exhibit a decreased quantum yield of ~0.04. The size-dependence of the quantum yield is likely due to the effect of particle size on the efficiency of energy transfer to various fluorescence quencher species present in the nanoparticle, consistent with both the results of exciton diffusion and energy transfer simulations and the size-dependent energy transfer efficiency we have previously observed in dye-doped CPdots.24
The effect of particle size on exciton-phonon coupling may also play a role.25
Figures of merit evaluating the multicolor conjugated polymer dots as fluorescent tags
The photostability of the CPdot nanoparticles is critically important for many fluorescence-based imaging applications, particularly for long-term imaging and tracking experiments.26
The photostability of a fluorescent dye or nanoparticle can be characterized by photobleaching quantum yield (ϕB
), which is the number of molecules that have been photobleached divided by the total number of photons absorbed over a given time interval. Typical fluorescent dyes exhibit photobleaching quantum yields in the range of 10-4
and exhibit single exponential photobleaching kinetics under low excitation intensity. The photobleaching of conjugated polymers is more complicated due to the complex set of interactions involving a large number of species such as excitons, polarons, molecular oxygen, and fluorescence-quenching sites of unknown structure.27-29
Photobleaching kinetics of the aqueous CPdot suspensions were observed to vary substantially from one polymer to another (). While PFO and PFPV dots exhibit single exponential photobleaching decays, the photobleaching kinetics of MEH-PPV dots contain a fast component and a slow component. The PPE dots exhibit unusual photobleaching behavior: initial light excitation increases the fluorescence quantum yield, resulting in a rapid increase in fluorescence followed by slow photobleaching. The PFBT dots appear to be remarkably photostable — photobleaching for two hours does not result in observable decrease in fluorescence intensity. These observations indicate complex photophysics in these nanoscale multichromophoric systems, and more detailed investigations are needed to better understand how these phenomena are determined by the polymer structure and environment. An estimate of both the photobleaching quantum yield and the photon number (total number of photons that a particle emits prior to photobleaching) can be obtained from the rate constants obtained from the photobleaching kinetics measurements.6
As shown in , the photon numbers of the CPdots are 3~4 orders of magnitude larger than those of typical fluorescent dyes,6
indicating promise for long-term imaging and tracking applications.
For high speed applications such as flow cytometry and high-speed imaging and tracking, a key figure of merit is the fluorescence radiative rate. Fluorescence decay kinetics () were obtained using the time-correlated single photon counting technique (TCSPC) and excited state lifetimes were extracted from the kinetics traces using custom software. All decay traces of the CPdots (with the exception of the 480 emission band of PPE dots) can be fit adequately with a single exponential function, and the lifetime results are listed in . The lifetimes of PFPV and MEHPPV dots were determined to be ~130 ps, while β-phase PFO and PFBT dots show longer lifetimes around ~270 ps and ~600 ps, respectively. PPE dots display complex fluorescence decay kinetics, the 440 nm emission peak shows a single exponential decay with time constant of 242 ps, while the 480 nm emission exhibits a bi-exponential decay with time constants of 276 ps and 1.56 ns. This observation is similar to previously observed fluorescence decay kinetics of PPE thin films, which are characterized as a heterogeneous system containing ordered and disordered polymer chains.23
The fluorescence radiative rate constant kR
and non-radiative rate constant kNR
were estimated by combining the quantum yield [ϕ
] and fluorescence lifetime results [τ
]. Typical fluorescent dyes exhibit fluorescence radiative rates on the order of ~108
. As shown in , the CPdots exhibit a fluorescence radiative rate (108
) similar to or somewhat higher than that of typical fluorescent dyes, while single quantum dots emit at rates about two orders of magnitude lower.30
The fluorescence radiative rates of the CPdots are at or above those of other fluorophores used in flow cytometry and imaging.
Single particle fluorescence imaging and kinetics studies provide further evidence of the CPdots as fluorescent probes for single molecule applications. Photobleaching studies of single PFBT dots (particle size: 10±3 nm) dispersed on a glass coverslip were performed using a custom built wide-field epifluorescence microscope. Bright, near-diffraction-limited spots corresponding to individual PFBT dots were observed (). Single particle photobleaching measurements were obtained by acquiring a series of consecutive frames. The number of fluorescence photons emitted per frame for a given particle was estimated by integrating the CCD signal over the fluorescence spot, and then scaling that value with the detection efficiency and amplification factor of the detector and the overall collection efficiency of the microscope. The photobleaching trajectories could be roughly categorized into two types exemplified by the curves shown in . Most of the particles exhibit continuous photobleaching behavior with no observable fluorescence blinking, as indicated by the green curve, while some blinking was often observed in dimmer particles (blue curve). This is consistent with the size dependent blinking observed in other conjugated polymers:19,31
the fluorescence of the smaller (dimmer) particles fluctuates due to the small number of emitting chromophores and the reversible on-off dynamics resulting in sizable fluctuations in the fraction of chromophores in the “on” state, while larger particles (> 10 nm dia.) result in relatively steady fluorescence there are contributions from a larger number of chromophores, resulting in smaller fluctuations. As fluorescent probes for imaging or single particle tracking, the steady fluorescence of the larger CPdots compares favorably to that of conventional dyes and quantum dots, which often exhibit pronounced blinking on time scales of microseconds to hundreds of seconds,12
although such blinking can be suppressed in some cases.32
Statistical analyses of multiple photobleaching trajectories indicate that approximately ~6×108
photons per particle (~10 nm diameter) were emitted prior to photobleaching (). The average photon number (number of photons emitted prior to photobleaching) obtained from single particle measurements is roughly consistent with the bulk photobleaching results (given an estimated uncertainty of ~20% in the determined photon number). Additional photobleaching experiments performed under nitrogen resulted in photon numbers roughly two orders of magnitude higher, indicating that oxygen is likely involved in the photobleaching mechanism (data not shown).
Figure 3 (a) A 10 μm×10 μm fluorescence image of single PFBT dots immobilized on a glass coverslip. (b) Photobleaching trajectories of single PFBT dots. No obvious blinking was observed for larger PFBT dots (green), while it was sometimes (more ...)
In single molecule experiments and other measurements involving high excitation intensities, the emission rate is often determined by saturation due to the presence of triplets and other long-lived nonfluorescent species. Typical fluorescent dyes exhibit a saturated emission rate on the order of 106
due to triplet saturation.33
This picture is complicated for multichromophoric systems such as conjugated polymers, as evidenced by photon antibunching studies and single molecule blinking studies that indicate roughly between 1 and 3 independent emitters for single conjugated polymer chains.34, 35
This value is much lower than the chromophore number that is expected based on the extinction coefficients and conjugation lengths of conjugated polymers, and the discrepancy can be attributed to funneling of excitons to a small number of emitters. Fluorescence saturation studies of single PFBT dots were performed under nitrogen protection so that photobleaching was negligible (however, the absence of oxygen also tends to increase triplet lifetimes, resulting in increased triplet saturation36
). As shown in , the fluorescence signal of a PFBT particle shows power saturation behavior that is well described by the saturation equation R
)(1 + I/Is
, where I
are the excitation and the saturation intensities, respectively. From the fit of this particle, we obtain a saturation emission rate R∞
= 0.5 kWcm-2
. Statistical analysis of several particles yielded saturation emission rates ranging from 107
. The mean saturation emission rate is roughly 100 times higher than that of typical molecular dyes, and at least 3 orders of magnitude higher than that of colloidal semiconductor quantum dots. The single particle saturation emission rate is markedly lower than the expected per-particle radiative rate based on the results of prior single molecule studies and the number of polymer chains per nanoparticle (~40). This apparent discrepancy is likely due to the larger number of available pathways for exciton diffusion in the nanoparticles relative to single conjugated polymer molecules, resulting in additional energy funneling.
The high absorption cross-sections, bright fluorescence, and large photon numbers of the CPdots indicate great potential for single molecule imaging and tracking in living cells. As compared to membrane-permeable organic dyes, the use of nanoparticles for intracellular imaging is generally complicated by the challenge of delivering the nanoparticles to the interior of the cell.37
A variety of strategies including invasive methods such as electroporation and microinjection have been demonstrated for intracellular delivery of inorganic quantum dots.38
Here we demonstrate the effective use of CPdots as an extremely bright fluid phase marker of pinocytosis in J774.A1 macrophages. The selection of the cell system is based on the ability of macrophages to efficiently ingest cellular debris, pathogens, and small particles such as CPdots. shows differential interference contrast (DIC) images and fluorescence images of these mouse macrophage-like cells that had been incubated with PPE, PFPV, PFBT, and MEH-PPV dots, respectively. These images clearly indicate internalization of the CPdots by the cells and show a staining pattern consistent with other widely used fluid-phase markers such as an organic dye conjugated to a high molecular weight dextran. Consistent with fluid-phase uptake of CPdots, these representative images show perinuclear labeling and brightly fluorescent vacuoles and organelles (e.g.
pinosomes and lysosomes). Preliminary co-localization studies with Texas Red dextran and Lysotracker Red favor fluid-phase uptake as the most likely primary uptake mechanism. More diffuse nanoparticle fluorescence was observed in the cytoplasm, possibly indicating that some of the CPdots crossed cell membranes. The CPdots do not appear to exhibit appreciable cytotoxicity under the current incubation time and loading concentration. The nanoparticles also appear to be stable (no evidence of degradation) in cell growth medium. A detailed understanding of the factors affecting nanoparticle uptake as well as the fate of the nanoparticles requires additional investigation. In addition, imaging or tracking specific intracellular species will require targeting of the CPdots via encapsulation and bioconjugation.
Differential interference contrast (DIC) images (top), and fluorescence images (bottom) of macrophage cells labeled with macrophage cells labeled with PPE, PFPV, PFBT and MEHPPV dots, respectively. Scale bar: 10 μm.