«By Yusuf Nur A thesis submitted to The University of Birmingham for the Degree of DOCTOR OF PHILOSOPHY School of Geography, Earth and Environmental ...»
3.2) was observed after heating the solution to 70 oC and waiting about 30 minutes. The total reaction time of these last samples was 3 hours. This slow change can be associated with the weak reducing effect of the PVP. Apart from the duration of the methods, there are a number of experimental conditions between the methods which needs to be considered. High temperature of around 70oC is needed for PVP hot method followed by a extensive cleaning process with water-ethanol solution (3:1 ratio) for the removal of the excess PVP. For the other two methods the synthesis was conducted at room temperature with no further washing process. Practically, PVP hot synthesis (Chapter 4: section 126.96.36.199) method lasted longer, required higher temperature, more chemicals were used for cleaning and resuspending step and the result was less monodisperse than the other two synthesis methods as can be seen in Table 5-1 below.
5.2.2 Characterisation of AuNPs For the characterisation of synthesised NPs, their main physicochemical properties have been measured and analysed. Among the properties which have been measured are size, shape, surface plasmon, monodispersity, surface charge, dissolution and stability in different environmental relevant conditions.
188.8.131.52 Presence of gold nanoparticles
In all three different synthesis methods summarised above and described in Chapter 4:
section 4.2, the colours of the solutions have changed from yellow, due to gold ions from the precursor to ruby red which is a characteristic for the formation of gold material in nanoscale (see Figure 5-1 below).
In PVP cold synthesis method (see above) and in citrate method, the
approximately 30 minutes.
Figure 5-1: Yellow colour of the gold solution changed into ruby red during the formation of AuNPs.
The colour change manifested is caused by surface plasmon resonance (SPR) of the forming gold NPs. Therefore, the presence of gold NPs in the final suspensions was further confirmed by Uv-vis absorbance measurement which shows the NPs localised surface plasmon resonance (LSPR) behaviours. The spectra are given in Figure 5-2 and clearly indicate the characteristic sharp absorbance peak around 520 nm, which is a characteristic of gold (Daniel and Astruc, 2003, Liz-MarzÃ¡n, 2005). The narrow peak and absence of other peaks above 550 nm indicates the monodispersity of the particles.
1 0.8 0.6 0.4 0.2 0 350 400 450 500 550 600 650 700
Figure 5-2: The surface plasmon resonance (SPR) of freshly prepared AuNPs and the maximum absorption takes place around 520 nm.
184.108.40.206 Size and size distribution The size and size distribution of the NPs have been studied using a number of different but complementary imaging and size measurements techniques. Among the techniques used for the size measurements include dynamic light scattering (DLS), field flow fractionation (FFF), transmission electron microscope (TEM), atomic force microscope (AFM) and ultraviolet visible spectroscopy (Uv-vis). Detailed description of the practical application and the theoretical background of each technique were given in sections 4.4 and Chapter 3: of the
220.127.116.11.1 DLS and FFF measured hydrodynamic sizes DLS and FFF were used to quantify the hydrodynamic diameter which is the core size of the particles plus the size of the coating agents. For the samples measured, the sizes from the two techniques were fairly comparable as can be seen in Table 5-1.The hydrodynamic z-average values, measured with DLS, of samples vary from 11.55 nm to 41.77 nm for citrate stabilised samples and from 22.60 nm to 102 nm for PVP stabilised samples. All DLS data plus standard deviations of the measurement were recorded in Table 5-1. DLS diagrams are presented in Figure 5-4 and it shows that samples synthesized have different sizes. Figure 5-4 also shows that most of the samples are monomodal in distribution except the samples synthesised through PVP hot method which shows two distinct peaks for both measured samples (G7 and G8). Figure 5-3 represents the FFF raw data of the three standards showing their relative elution times and the UV_vis detector signal by 254 nm wavelength.
Figure 5-3: Raw data of the three standards as measured with FFF. The bigger the size of the nanoparticles the longer it takes for NPs to get eluted through the FFF column.
The sizes of the samples measured with FFFF show similar range as the DLS size. G2 has
difference between the sizes measured with DLS and FFF is in the margin of error. For sample G2, T-experimental (1.27 calculated from the samples means and the standard deviation) is smaller than T-critical 2.45 (degree of freedom 11) and for sample G5, Texperiemtal is 1.18 than T-critical 2.45. Thus they are not significantly different.
Figure 5-4: DLS diagrams of eight samples illustrating the sizes of the particles. Samples G1.G2 and G3 are stabilised by citrate, samples G4, G5, and G6 are reduced by NaOH and stabilised by PVP while samples G7 and G8 are stabilised PVP. The peaks around 100 nm for the last two samples are big gold particles which shows that the last two samples are clearly more polydispersity than the other samples.
Synthesised particles demonstrated colours varying from red for smaller particles to purple for larger particles as shown in Figure 5-5. The reason why NPs have different colours can be attributed to the fact that smaller particles (30 nm) absorb the blue section of the white light spectrum and reflect the red portion which is responsible for their red colour. On the other hand the larger particles due to their size, absorb larger wavelength in the red portion of the spectrum and reflect the blue section which explains why they have purple/blue colour. This property can be used to study the kinetics of size changes phenomena such as agglomeration and aggregation of the particles (de la Rica et al., 2012, Hayden et al., 2012).
Figure 5-5: Freshly synthesised AuNPs of different sizes. Ruby red colour at the left is for smaller (around 10 nm) AuNPs while the purple colour at the right is manifested by bigger particles (around 45 nm hydrodynamic size).
For further analysis of the particles with different sizes, their surface Plasmon resonance (SPR) properties were recorded and compared among them using modern double beam Uv_vis
the electromagnetic waves including visible light and the delocalised surface electrons of the metallic nanoparticles. The position of the maximum absorbance measured with Uv-vis depends on a number of factors including the shape, the dialectric constant, the surrounding media and the size of the nanoparticle s(Eustis and El-Sayed, 2006).
SPR spectra of AuNPs with range of sizes were recorded and presented in Figure 5-6 a below. Then the hydrodynamice diameters measured with DLS were plotted against the wavelengths corresponding to the maximum absorbance of each particle size and illustrated in Figure 5-6 b below. There is understandable positive correlation between the two variables: the size and the wavelength of the maximum absorbance, with regression constant ( R2 = 0.94). Here, Uv-vis data has supported the fact that the synthesised gold nanoparticles have a range of different sizes. It can be clearly seen that the particles with bigger sizes have their maximum absorbance at a relatively higher wavelength toward the red section of the visible light spectrum. The wavelength of the maximum absorbencies of the NPs varied from 517 nm to 529 nm when the z-average values size of the hydrodynamic diameter of particles measured with the DLS varied from 11 nm to 92 nm as can be seen in Figure 5-6 below.
To test the significant of the correlation, t-value was calculation using equation Equation 4-8 in
section 4.6 Chapter 4:.
Then tdist(t,freedom,tail) function of the excel software was used to calculate probability value of
4.5E0-8 which much smaller than 95% significance value of 0.05. Thus the correlation is notsignificant.
529 1.2 524 523 1.1 522 1 520 0.9
Figure 5-6: Graph showing the relationship between the size and the SPR maximum. The bigger the size of the particles the higher the wave length of the maximum absorption Part a) illustrates the red shift of the maximum SPR due to the increasing size. Part b) presents the positive correlation between z-average and the SPR maximum.
TEM micrographs of the synthesised particles were measured using TEM jeol1200 (see Chapter 4:
2 for the imaging procedure plus sample preparation methods and section 3.1.2 for theoretical background of the TEM technique) and the microgarphs of representative samples are presented in Figure 5-7 for citrate and Figure 5-8 for PVP capped gold samples. Size distributions histograms of the different types of the synthesised gold nanoparticles were also calculated and shown in these graphs. Variations in size were achieved with varying the initial molar ratio between the tetrachloroauric acid and the capping agents (citrate and PVP) (Frens, 1973) (see Table 4-1 in Chapter 4: for the experimental condition and the concentration of the reactants used for the synthesis of the particles).
The average core sizes of citrate capped particles from a minimum number of 100 particles were found to be in the range 7.1 nm to 32.6 nm while PVP-coated particles (minimum 100 particles) have core sizes varying 11.4 nm to around 85 nm (see details with standard deviation in Table 5-1).
Figure 5-7: TEM images and size distribution histograms of citrate capped AuNPS of two sizes of AuNPs. G1 particles are clearly bigger in size than in G2. See full description of the nanoparticles in Table 5-1.
50 nm Figure 5-8: TEM images and size distribution histograms of cold method prepared PVP capped AuNPs of different sizes. Righthand side of the figure is presented with excel calculated size distribution of the particles, For further analysis of the size of the synthesised gold nanoparticles, samples were immobilised on mica sheet (see section 4.4.3 of Chapter 4: for details of procedures) and AFM topographs were recorded using noncontact mode of the AFM. Average diameter and size distributions of the particles were calculated and illustrated in Figure 5-9 below and recorded in Table 5-1.
Figure 5-9:Topographs of three different particles as measured with AFM and the corresponding size distribution. AFM images courtesy Dr Mohamed Baalousha.
The size measurements results of the abovementioned techniques were summarised in Table 5-1 below. In addition to the size data, the table gives standard deviation values and other relevant technique specific parameters such as polydispersity index (PDI) for DLS data.
Table 5-1: Hydrodynamic diameters and core sizes of AuNPs as measured with different techniques.
The data in Table 5-1 shows that the DLS-measured Z-average hydrodynamic diameters of the synthesized particles are always larger than the sizes obtained by TEM and AFM analysis and are fairly comparable with FFF data which also gives a measurement of hydrodynamic diameter calculated from the diffusion coefficient of the particles. To see whether the abovementioned difference is significant, student ttest ( P = 0.05) was conducted and the results for citrate capped samples and PVP capped samples were summarised in the following Table 5-2 below. The averages
calculated from at least 100 particles. To find the value of T-critical from statistical tables, the size of the smaller DLS replicates (5 replicates) were used. For nearly all samples analysed texperimental were larger than the t-critical which means that the difference is significant. This difference between DLS and FFF on one hand and TEM and AFM on the other hand, can be caused by the combination effect of a number of factors including the degree of polydispersity of the particles and type and size of the coating agents. In the sense that the higher the polydispersity the higher the difference between the DLS and TEM/AFM sizes. Regarding coating agents, the organic stabilizer in this case citrate and PVP is transparent to electrons, so TEM gives the size of only more electron dense metallic core of the particles, whereas DLS measures the hydrodynamic diameter of the gold NP calculated from the diffusion coefficient. The DLS size, therefore, includes the core, the capping agent and any associated water (Diegoli et al., 2008). Another important point is that the difference between DLS and TEM is always higher for the PVP stabilised samples than the citrate stabilised samples (see Table 5-1). This is due to the fact that PVP is a long chain polymer while citrate is a much shorter molecule.
Table 5-2: Significant test data for the size difference between TEM core sizes and DLS hydrodynamic diameters.
TEM and AFM images of the freshly synthesised NPs presented in Figure 5-7 and Figure 5-8 above respectively have clearly manifested the spherical shapes of the synthesised nanoparticles. The images clearly show the presence of individual spherical particles in the samples. Quantification of the shapes of the all particles was attempted by calculating their shape factors using Equation 4-1 in Chapter 4: materials and methodology. This equation gives information about the sphericity of the particles. Perfect spherical particles will have shape factor values of 1 while other particles will have values less than 1. The nearer the value to 1 the more round the particle is.