The Influence of Interfacial Effects by PVDF on the Fluorescent Properties of Rhodamine 6G

In Chapter 1, Fluorescent enhancements have been achieved using a simple layered structure: fluorophore, polymer/metal ion, glass substrate. The polymer/metal ion layer apparently has a strong influence on the emission response of the fluorophore by removing the dye aggregation. This data supports that the addition of higher concentration of hydrated transition metal salt increases the production of β-phase in the Polyvinylidene Difluoride (PVDF). The absorption spectra intensity increased as the amount of Zn is increased in the substrate while the Rhodamine 6G (Rh6G) thickness is kept constant. Investigation into the means of β-phase production and the influence of the interfacial region effect on the fluorescence enhancement was completed and reported in this work. The goal of this study is to understand the interfacial properties that control the nature of the fluorescent emission and determine the structure of the fluorophore on different substrates. Chapter 2 of this dissertation shows an investigation of the role of the polymer substrate, the solvent and the reaction between Rh6G and Zn. Studying the reaction between Rh6G and Zn was done on glass slides in the absence of polymer. In order to study the role of the solvent, films of PVDF doped with Zn were cast from pure acetone. PVDF polymer was replaced by PMMA to evaluate the role of the polymer in the films. The response of emission signal of TNT, as a function of Rhodamine 6G concentration and mol % of Zn has been studied in Chapter 3. Using a different concentration of Rhodamine 6G and mol % of metal ion, the signal is quenched by a notably large amount. Different analytes can be applied and emission signals can be collected to find a pattern that could be used to identify the explosives.


INTRODUCTION
We have been investigating surface properties of fluorescent dyes since they have been shown to be effective sensors for a number of analytes. [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15] We were able to demonstrate that the emission intensity can be significantly increased, up to three orders of magnitude, using a three layer structure where a transparent polymer was sandwiched between the fluorophore and the substrate 11 , but the origin of this enhancement was unclear. Fluorescent spectroscopy is often used to study surface properties. [16][17][18][19][20][21][22] So, as a representative example we used rhodamine 6G (Rh6G), whose properties have been heavily studied, 23-31 as a surface probe on a variety of substrate structures to determine the cause for the emission enhancement. When Rh6G is deposited as a single layer on a glass surface the fluorescent properties are strongly thickness dependent. 32 For thin films, submonolayer thickness, the emission is dominated by monomers of Rh6G. However, as the thickness increases both excimer and exciton emission could be detected. When the film is several nm thick, representing more than three layers of Rh6G, the emission intensity is significantly quenched and arises primarily from excitons. When a layer of poly(vinylidene difluoride) (PVDF) is placed between the glass substrate and the Rh6G surface the overall emission is enhanced. 11,33 Three components contributed to the enhancement.
First, the PVDF surface was rough, leading to deposition of more fluorophores per unit area on the polymer. Second, the polymer layer allowed internal reflection of the incident light beam, which led to increased absorption and subsequent emission.
Finally, the polarity of the polymer may have contributed to the increased emission.
However, structurally the Rh6G followed the same pattern as on glass: at low thicknesses monomers were dominant and at high thicknesses aggregation determined the photophysical properties.
PVDF is an interesting polymer because it can be found primarily in three different phases: the nonpolar α-phase, the polar γ-phase, and the ferroelectric βphase. [34][35][36][37][38][39][40][41][42][43] Casting films of PVDF into uniform phases is challengingnormally a mixture of α-and γ-phases is found from typical spin-casting conditions. 38 We recently reported that doping PVDF with a transition metal salt could induce β-phase formation. 44 In that work Co 2+ was used as the doping agent and we showed that the complexation of the metal ion to the F atoms in the PVDF induced a cooperative effect to increase the amount of β-phase. Thin films of the Co 2+ -doped PVDF behaved as relaxor ferroelectrics, indicating that the electrical properties of the films were controlled by the β-phase composition.
In this work we use Zn 2+ as a dopant to control the phase behavior of PVDF in order to study the effects on Rh6G at the polymer/fluorophore interface. We chose zinc(II) over cobalt(II) to remove the complications of the d-d transitions in the visible absorption spectrum and to eliminate any effects that might be induced by the magnetic moment of Co 2+ . As expected, the zinc ion induced β-phase formation.
However, the presence of the metal ion also completely suppressed aggregation of Rh6G on the polymer surface, even at a doping level as low as 1 mol % Zn 2+ .

EXPERIMENTAL
Polyvinylidene fluoride with M w = 534,000 g/mol was purchased from Sigma Aldrich. Zinc (II) nitrate hexahydrate, spectral grade acetone, and N, Ndimethylformamide (DMF) solvent were purchased from Fisher Scientific. Glass slides were used as a substrate for film formation. The glass slides were cut into dimensions of 3.75 cm × 1.75 cm. The slides were placed in a container of 95% ethanol (EtOH) and sonicated for 15 minutes. Then they were rinsed three times and placed in distilled water and sonicated for an additional 15 minutes, then dried with N 2 gas. 13 A solvent system of 90/10 v/v of acetone to DMF was used to dissolve the PVDF at a 3 % w/v ratio. 6,14 The polymer solution was placed in a Branson 3510 ultrasonic cleaning device for 3 hours at 40 o C to ensure that all of the polymer dissolved. Specified mole percentages of zinc(II) determined the mass of Zn(NO 3 ) 2 ·6H 2 O that was added to the polymer solution after the sonication process, using a mechanical swirling technique to dissolve the salt followed by an additional 2 to 3 minutes sonication time until the salt was fully dissolved in solution. Nine different Zn 2+ mole percentages ranging from 1 to 5 mol % were prepared. A series of Rh6G concentrations ranging from 1×10 -4 M to 9×10 -4 M were used.
A Laurell Technologies WS-400B-6NPP/LITE spin-coater was used to prepare films on the glass slides. 300 µL of polymer solution was placed on the substrate and then spun at 1200 rpm for 45 s with an acceleration of 1080 s -2 . After spin casting of polymer solutions was complete, the samples were then placed in an oven set to 60 o C for 2 min to dry. The Rh6G solutions were also spun-cast in a similar manner. A 6 volume of 50 µL was placed on the polymer surface and spun at 1200 rpm for 45 s with an acceleration of 1080 s -2 .
A Perkin-Elmer Lambda 1050 spectrometer was employed for obtaining UV-Vis spectra. Before the absorbance spectrums were collected, a blank of a clean glass slide was taken. A Horiba (JobinYvon) Fluorolog spectrometer was used for steadystate fluorescence measurements and data were collected at 60-degree angle relative to the excitation beam. The emission wavelength range was from 520 to 800 nm with a slit width of 3.0 nm. The excitation wavelength range was from 300 nm to 800 nm with a slit width of 2.0 nm to correspond with the absorbance spectrum. A Perkin-Elmer Spectrum 100 FTIR spectrometer was used for infrared measurements in ATR mode. Samples were scanned from 650 to 4000 cm -1 at a resolution of 1 cm -1 . A Filmetrics F40 microscope was used to measure the film thickness. TGA measurements were done on a TA Instruments Q50 between 25 and 400 o C with a heating rate of 10 o C/min. An Agilent Technologies 5500 AFM was used to collect images of the surface using tapping mode.

RESULTS AND DISCUSSION
In this study, spin casting is used to deposit the polymer onto the glass substrate and potentially three phases may exist in the thin films of PVDF. FTIR spectroscopy can easily identify the different phases of PVDF, 38 and Figure 1 shows the changes in the spectra as Zn 2+ is added to the PVDF film. The α-phase can be identified by a peak at 764 cm -1 , the β-phase by peaks at 840 cm -1 and 1275 cm -1 , and the-γ phase by peaks at 840 cm -1 and 1233 cm -1 . The FTIR spectroscopy showed an increase in the amount of β-phase and g-phase present and a decrease in the amount of !-phase in the thin films as the amount of Zn 2+ is increased, as shown in Fig. 1A. In the carbonyl region there is a strong feature that is not associated with PVDF, as shown in Fig. 1B. The increasing intensity around 1650 cm -1 is due to the presence of coordinated DMF, which increases as the amount of Zn 2+ increases. However, even in the absence of Zn 2+ a small amount of DMF is retained by the PVDF, as indicated by the solid black line in Fig. 1B. Casting films from pure acetone with no DMF removes the carbonyl peaks but there also is a much smaller amount of β-phase formed.
Thermal Gravimetric Analysis (TGA) was used to help confirm the role of the DMF identified from the FTIR spectroscopy. Figure 2A shows that as the amount of zinc ion is increased there is an increasing mass loss between room temperature and ~230 °C. The broad feature from room temperature to ~180 o C is assigned to loss of H 2 O. The wide temperature range suggests that the water is not coordinated to the metal ion but is trapped in the polymer during spin-casting. The sharper mass loss centered at 210 o C is assigned to loss of DMF that is coordinated to the Zn 2+ ion. The small features above 300 o C are assigned to decomposition of nitrate ion. Figure 2B shows the quantitative fit of the mass loss data. The fit arises from assigning 6 coordinated DMF molecules to each Zn 2+ ion, another 0.08 DMF molecules for each repeat unit in the PVDF, and 6 H 2 O molecules per Zn 2+ . This is consistent with the FTIR results, which show a small amount of DMF in the polymer even in the absence 9 of zinc(II) ion. The water stoichiometry matches what was used in the sample preparation, the hexahydrate of zinc nitrate.   and 4C are all cast with the same concentrations of Rh6G, so presumably are of comparable thicknesses, but as the Zn 2+ level in the film increases the absorbance also increases. Since the Zn 2+ ion affects the absorbance intensity the nominal thickness of the Rh6G cannot be determined from the absorbance spectra, as was done previously. 32,33 Directly measuring the thickness of the Rh6G film by other methods also proved unsuccessful because the surface roughness of the polymer film interferes.
As the films get thicker there is no shift in wavelength maximum, in contrast to what 11 is observed for Rh6G cast onto glass substrates 32 and undoped PVDF thin films. 33 In the absence of Zn 2+ the Rh6G spectra show a gradual shift in λ max from ~520 nm to ~560 nm as the Rh6G gets thicker, which is attributed to aggregation of the dye molecules on the surface. [23][24][25][26][27][28][29][30][31][32][33] In the presence of the Zn 2+ , all the spectra have maximum wavelengths ~520 nm. There also is a shoulder at ~495 nm, which is assigned to an exciton. 32,33 Interaction of the Zn 2+ ions with the Rh6G on the surface is preventing aggregation.  The Zn 2+ in the PVDF film also affects the spectral intensity, as shown in Figure 5. As the Zn 2+ concentration increases in the film the intensity of the absorption increases even though the surface coverage of the Rh6G should be constant. This indicates that the zinc ion is interacting with surface Rh6G in such a fashion to affect the absorbance intensity but not the absorption energy.

13
All of the absorbance line shapes are similar. All of the spectra were fit to three peaks and these results are given in Table 1. The peak maxima are similar to those reported previously. 32,33 The peak at 521 nm is assigned to the monomer absorption while the features at 496 nm and 534 nm are assigned to an exciton pair with an oblique geometry. Even for the thickest films there is no evidence of a peak in the 560 nm region, which would be indicative of aggregation of the Rh6G on the surface. Likewise, at constant Zn 2+ doping the changes are 13 nm at 1 mol% and 40 nm for 5 14 mol%. The changes in the emission spectra are similar for either more Zn 2+ or more Rh6G and the two structural changes amplify the emission changes.
Similar behavior is observed for the total emission intensity, as shown in Figure 8. As more Rh6G is added to the surface the total emission intensity increases, which makes sense since there is more fluorophore present on the surface. However, as the amount of Zn 2+ doped into the PVDF increases, the total emission intensity also increases, even at a constant amount of Rh6G.   Each of the emission spectra can be deconvoluted into three peaks. As shown in Figure 9, the highest energy peak is found at 536 nm and is fairly constant (±4 nm) from low to high Zn 2+ levels or from low to high Rh6G levels. At low Rh6G coverage, the other two peaks (~560 nm and ~600 nm) are only slightly affected by the concentration of the Zn 2+ in the underlying polymer layer. However, as the coverage of the Rh6G increases, the influence of the Zn 2+ also increases so that at the highest Rh6G thickness the peaks shift from ~570 nm to ~615 nm (ω 2 ) and ~605 nm to ~675 nm (ω 3 ).    giving three lifetime parameters. The shortest lifetime was always less than 0.5 ns and is assigned to scattering from the film. The other two lifetimes are associated with the Rh6G and are given in Table 2. The shorter lifetime is τ 1 = 2.4±0.4 ns is assigned to relaxation from the Rh6G isolated monomer. In dilute DMF solution the lifetime of Rh6G is 4.2 ns, 10 which indicates that in the thin film the underlying substrate provides more nonradiative pathways than in solvent. The longer lifetime, τ 2 = 10.4±1.6 ns, is assigned to de-excitation from aggregated areas of the thin film. Since there is no direct absorption measured for aggregates, the longer lived excited states must arise from energy transfer or charge transfer from the monomer or exciton.
Another source of this longer lifetime could be exciton diffusion within the Rh6G thin film. Figure 11. Excited state decays for Rh6G on Zn 2+ -doped PVDF. IRF = instrument response function.
As summarized in Table 3, the presence of the Zn 2+ ion in the PVDF film has a profound effect on the photophysics of Rh6G. In the absorption spectra, the intensity of absorption increases for both increased Rh6G thickness and increased Zn 2+ .
However, absorption associated with aggregation is completely suppressed. An electric field created by the Zn 2+ could increase the transition moment of the monomer but why this would eliminate aggregate absorption is not known. In the emission spectra the Zn 2+ again has important effects. As Rh6G or Zn 2+ increases, the spectral intensity increases and the maxima shift to lower energy. there is a significant shift of the emission maxima to lower energy, revealing a high energy emission not previously observed. Three of the emission features can be assigned to the monomer excited state and the two exciton excited states. The lowest energy emission must be either from aggregates populated by energy transfer or excitons that diffuse into the Rh6G thin film. Since the overlap of the absorption spectra and the emission spectra is quite small (see Fig. 10), exciton diffusion seems more likely, which is also consistent with the measured lifetimes.

INTRODUCTION
Fluorescence spectroscopy is a very powerful tool that has been used in many areas, such as analytical chemistry, biology and sensors. Detecting explosives is one of the most challenging applications for fluorescence sensing !!! , which has been an interest for several years. !,!!! In this work, Rhodamine 6G is used as the fluorophore because it exhibits some interesting chemistry. Rhodamine 6G has well known optical properties. However, when working with solid phase dyes, the geometry of the layered Dye molecules aggregation is a major source of decreasing the fluorescent emission. !!!" Therefore, in order to achieve high fluorescence intensity, a material of dye monomers should be prepared. To begin understanding the role of the transition metal salt on increase the absorbance, enhance the fluorescence emission and remove rhodamine 6G aggregation, the absorption and emission spectra of Rh6G on PMMA and PVDF cast from acetone were examined in more detail.
In chapter one, when PVDF polymer cast from (acetone/DMF) was doped with a few mole percent of zinc nitrate hexahydrate, the character of the absorbance spectra changed. As the films get thicker we do not see the shift in wavelength as in Rh6G on glass substrate and PVDF thin films. All the peaks have ~520 nm maximum wavelengths with a linear relationship between the concentration of Rh6G and the zinc mol %. At low Rh6G concentration there are two features, a peak at ~520 nm and a shoulder at ~480 nm. At high Rh6G concentrations the spectra have the same line shape with no peak growing at 550 nm. The same effect is seen independent of the Zn !! concentration, but the absorbance is significantly higher. This indicates that the presence of Zn !! is stopping Rh6G from aggregating to form the new absorbing species centered at 550 nm. However, TGA and FTIR measurements also showed significant DMF retained in the films, coordinated to the Zn !! .
The work reported here is to investigate the role of the polymer substrate, the solvent and potential in the actions between Rh6G and Zn !! . Determination of a reaction between Rh6G and Zn !! is explored using absorption spectroscopy in the absence of a polymer. Next, films of PVDF doped with Zn !! and cast from pure acetone to find the role of DMF. Finally, PMMA is used as the polymer substrate to evaluate the role of PVDF (Fig.1).

Film Materials
Poly (methyl methacrylate) (PMMA) with a molecular weight 100.12 g/mol was purchased from Sigma Aldrich. Polyvinylidene fluoride with a molecular weight 534,000 g/mol was purchased from Sigma Aldrich. Zinc (II) nitrate hexahydrate salt with a molecular weight 297.47 g/mol was purchased from Fisher Science Education.
Spectral grade acetone and N, N-Dimethylformamide (DMF) solvents were purchased from Fisher Scientific. Glass slides were used as a substrate for film formation.

Sample Preparation
The glass slides were cut into dimensions of 3.75 cm x 1.75 cm. The slides were placed in a container of 95% ethanol (EtOH) and sonicated for 15 minutes. Then they were rinsed three times and placed in distilled water and sonicated for an additional 15 minutes, then dried with N ! gas. !" A solvent system of 90/10 volume /volume of acetone to N, N-Dimethylformamide (DMF) was used for dissolution of 3% weight /volume ratio of PVDF polymer to overall volume of solvent. !

MEASUREMENTS
Perkin Elmer Lambda 1050 UV-Visible spectrometer was employed for obtaining UV-Vis spectra. Before the absorbance spectrums were collected, a blank of clean glass slide was taken. A Horiba (JobinYvon) Fluorolog spectrometer was used for steady-state fluorescence measurements and data were collected at a 60-degree angle. The emission wavelength range was from 520 to 800 nm with a slit width of 3.0 58 nm. The excitation wavelength range was from 300 nm to 800 nm with a slit width of 2.0 nm to correspond with the absorbance spectrum. The second hypothesis was the DMF solvation influence, in order to examine if that behavior was due to the DMF solvation, an experiment was conducted where a pure acetone was used as a solvent. It has been found from the optical absorbance spectra that using acetone solvation in PVDF/ Zn !! thin films doped with Rh6G has an absorbance maximum at ~520 nm and the shoulder on the red side at ~550 was removed ( Fig.2 and Fig.3).  However, the absorbance maxima has a very small different as a function of the zinc mol% unlike the DMF solvation. This experiment indicates the zinc metal ion is the causative factor of removing the dye molecular aggregation. Another fascinating result is that as the zinc ion increases in the film, the absorbance is increased as well at the same [Rh6G]. In the first place, this was attributed to the continuous enhancement in the maximum absorbance intensity to

RESULT AND DISCUSSION
Rh6G fluorophore orientation on the surface. It was though that in the presence of zinc metal ion the dye orientation would change facing the light, which would cause increasing in the absorbance intensity as the percentage of the metal ion was increased at the same concentration of Rh6G. However, the polarization experiment showed a random structure of the dye, since all of the peaks are overlapped no matter what value of polarization angle was used (Fig 4).

61
The third hypothesis was the presence of β -phase PVDF; to test this the PVDF polymer was replaced with PMMA.   5 shows an additional support to the major influence of zinc metal ions in removing the dye molecular aggregation. A single peak appears at ~550 nm with a shoulder at ~510 nm when PMMA thin film was doped with a high Rh6G concentration (9x10 !! M). However, doping PMMA/Zn !! with the same concentration of Rh6G has a single peak at ~530 nm.
To demonstrate that the emission enhancement was due to the presence of Zn !! and not the DMF solvation, an experiment was conducted using acetone solvation.  In Fig.6 an emission spectra of PVDF/Zn !! thin films was compared with a native PVDF films cast from acetone solvation. It shows that the presence of Zn !! has an increase in the emission intensity by 1.7x10 ! a. u. at high concentration of Rh6G.
This is another support to the hypothesis for the strong influence of zinc metal ions in removing the dye molecules aggregation, and enhances the emission intensity. Fig.7 shows an emission enhancement in the PMMA/Zn !! thin films compared with the native PMMA films cast from (DMF/Acetone) by 8.84x10 ! a. u. at high concentration of Rh6G, and this is in agreement with the previous hypothesis.

CONCLUSION
PVDF polymer cast from DMF/acetone mixture doped with a few mole percent of zinc nitrate hexahydrate increases the production of the ferroelectric βphase, emission intensity and the absorbance maxima. PVDF cast from acetone doped with Zn !! removed the aggregation. However, the absorbance maximum did not change.
PMMA polymer cast from DMF/acetone mixture doped with Zn !! increases the absorbance maximum with no aggregation. The previous result supports that the presence of Zn !! metal ion removes the aggregation. The DMF/Zn !! mixture increases the absorbance maximum as a function of Zn !! mole percent.
In a conclusion, Zn !! suppresses aggregation even at high [Rh6G] and emission enhancement still occurs. DMF causes absorption changes.
A number of experiments were conducted in order to understand the role of the interface on the fluorophore/polymer properties. UV-Visible spectroscopy was used to

INTRODUCTION
Developing a sensor device to detect explosives is an active area of research for the past decades. !!!" The reliable detection of explosives is an international concern and it is important for homeland security, environmental safety and the military. While there are many suitable methods for the detection of explosives, we have been using fluorescence as our technique because it has the benefit of sensitivity, low cost and frequently used for explosive sensors. An example of an effective explosive sensor was developed by Dr. Swager research group using a substrate with a conjugated fluorescent polymer. !" Since then an increasing interest has been paid to the detection of explosives using a fluorescent amplifying polymer technique. !",!" In chapter one, Fluorescent enhancements have been achieved using a simple layered structure: fluorophore, Polymer/metal ion, glass substrate. The polymer/metal ion layer had a strong influence on the emission response of the fluorophore by removing the dye aggregation. The achieved fluorescent enhancements by the layered structure can be used for the detection of explosives. !"!!" The idea behind fluorescent detection of explosives is measuring the emission signal intensity before and after applying the explosive to the sample: fluorophore (Rh6G), PVDF/Zn !! , glass substrate. The change in the signal intensity is either quenching or increasing. In our case the signal is quenching by a notably large amount. When the explosive molecule collides with Rhodamine 6G in the excited state, more nonradiative pathways become available. Rhodamine 6G loses its energy and the amount of emitted light decreases and that leads to a decrease in the fluorescent signal. !,!,!,!",!",!" Then the amount of signal loss or gain can be used to identify the analyte.
In order to find a pattern that can be used to distinguish the different   explosives, a table can be made with difference emission spectra after exposure to   various analytes: TNT, trinitrotoluene; TNB, trinitrobenzene; PETN, pentaerythritol tetranitrate; RDX, TATP, triacetone triperoxide; DNT, dinitrotoluene; NT, nitrotoluene. In this work, TNT, trinitrotoluene has been used as the analyte and a large quenching in the emission signal has been observed.

Film Materials
Polyvinylidene fluoride with a molecular weight 534,000 g/mol was purchased from Sigma Aldrich. Zinc (II) nitrate hexahydrate salt with a molecular weight 297.47 g/mol was purchased from Fisher Science Education. Spectral grade acetone and N, N-Dimethylformamide (DMF) solvents were purchased from Fisher Scientific. Glass slides were used as a substrate for film formation. Trinitrotoluene (TNT) was purchased from Sigma Aldrich.

Sample Preparation
The glass slides were cut into dimensions of 3.75 cm x 1.75 cm. The slides were placed in a container of 95% ethanol (EtOH) and sonicated for 15 minutes. Then they were rinsed three times and placed in distilled water and sonicated for an additional 15 minutes, then dried with N ! gas. !" A solvent system of 90/10 volume /volume of acetone to N, N-Dimethylformamide (DMF) was used for dissolution of 3% weight /volume ratio of PVDF polymer to overall volume of solvent.

MEASUREMENTS
A Horiba (JobinYvon) Fluorolog spectrometer was used for steady-state fluorescence measurements and data were collected at a 60-degree angle. The emission wavelength range was from 520 to 800 nm with a slit width of 3.0 nm. The excitation wavelength range was from 300 nm to 800 nm with a slit width of 2.0 nm to correspond to the absorbance spectrum.

RESULT AND DISCUSSION
In this section, the response of the emission signal of TNT, trinitrotoluene as a function of Rhodamine 6G concentration and mol % of Zinc (II) has been studied. At different concentrations of the fluorophore and mol % of metal ion, the signal is quenched by a notably large amount. This might be caused by the analyte interacting with the fluorophore to cause collisional quenching.  Increasing the emission signal after exposure to the analyte could be caused by the breaking of aggregates by the analyte.

CONCLUSION
In conclusion, the emission signal quenched by a notably large amount after adding TNT, trinitrotoluene to the layered structure: fluorophore, Polymer/metal ion, glass substrate. A table with different concentrations of Rhodamine 6G and mol% of Zinc(II) shows the decrease in the emission signal after adding the analyte. In future work, various analytes could be applied and emission signals could be collected to find a pattern that could be used to distinguish the different explosives.