**3. Results and Discussion**

### *3.1. Size and Elemental Composition of Particles*

The size of SiNPs was determined based on DLS measurement as shown in Figure 3. The average diameter of SiNPs was 330 ± 100 nm. The TEM image in Figure 4 shows similar SiNPs sizes.

**Figure 3.** Size distribution of silicone nanoparticles obtained from DLS. **Figure 3.** Size distribution of silicone nanoparticles obtained from DLS. **Figure 3.** Size distribution of silicone nanoparticles obtained from DLS.

**Figure 4.** TEM image of SiNPs. **Figure 4.** TEM image of SiNPs.

**Figure 4.** TEM image of SiNPs. The hydrolysis and condensation reaction of TEOS would produce silica particles based on the sol-gel process [27–29,34]. In this work, the condensation of TEOS occurred in the presence of PDMSdiol to result in silicone nanoparticles. The EDX results in Figure 5 showed that the elemental composition of SiNPs was consisted of 28.6% Si, 40.8% O, and 30.6% C. Considering that the atomic ratio of Si:O in PDMS is 1:1 while that in SiO2 is 1:2, the mole fractions of PDMS and SiO2 in SiNP can The hydrolysis and condensation reaction of TEOS would produce silica particles based on the sol-gel process [27–29,34]. In this work, the condensation of TEOS occurred in the presence of PDMSdiol to result in silicone nanoparticles. The EDX results in Figure 5 showed that the elemental composition of SiNPs was consisted of 28.6% Si, 40.8% O, and 30.6% C. Considering that the atomic ratio of Si:O in PDMS is 1:1 while that in SiO2 is 1:2, the mole fractions of PDMS and SiO2 in SiNP can be estimated by solving the following equations x + y = 0.286 for Si The hydrolysis and condensation reaction of TEOS would produce silica particles based on the sol-gel process [27–29,34]. In this work, the condensation of TEOS occurred in the presence of PDMS-diol to result in silicone nanoparticles. The EDX results in Figure 5 showed that the elemental composition of SiNPs was consisted of 28.6% Si, 40.8% O, and 30.6% C. Considering that the atomic ratio of Si:O in PDMS is 1:1 while that in SiO<sup>2</sup> is 1:2, the mole fractions of PDMS and SiO<sup>2</sup> in SiNP can be estimated by solving the following equations

$$\begin{cases} \text{ x} + \text{y} = 0.286 & \text{for Si} \\ \text{ x} + 2\text{y} = 0.408 & \text{for O} \\ \text{.} \quad \text{y} = 0.122, \text{ x} = 0.164 \end{cases}$$

⸫ y = 0.122, x = 0.164

Thus, the mole fraction of silicone is <sup>x</sup> <sup>x</sup>+<sup>y</sup> <sup>=</sup> 0.164 0.286 <sup>=</sup> 57.4% and that of silica is <sup>y</sup> x+y = 42.6%. In addition, the presence of carbon in SiNPs indicated that PDMS did incorporated into SiNPs. Thus, the mole fraction of silicone is x x + y = 0.164 0.286 = 57.4% and that of silica is <sup>y</sup> x + y = 42.6%. In addition, the presence of carbon in SiNPs indicated that PDMS did incorporated into SiNPs.

*Polymers* **2020**, *12*, x FOR PEER REVIEW 6 of 15

**Figure 5.** Micrograph of SiNPs from SEM-EDS. **Figure 5.** Micrograph of SiNPs from SEM-EDS.

Raman and FTIR were employed to examine the chemical structure of SiNPs synthesized from PDMS and TEOS. Figure 6 shows that the Raman spectrum of SiNPs exhibited strong PDMSs peaks appearing at 2909–2968 cm−1, 1409–1459 cm−1 (C-H groups), and 706–782 cm−1 (Si-C groups). These results were also found in the literature [35–37]. Furthermore, the peak at 483 cm−1 were attributed to Si-O-Si [38]. These characteristic peaks indicated that the reaction of PDMS and TEOS through the sol-gel process was successful. Comparing the spectra of TEOS, PDMS, and SiNPs in Figure 7, the peaks of Si-O-Si (1074 cm−1, 808 cm−1, 796 cm−1, 786 cm−1, 495 cm−1, 480 cm−1, 460 cm−1) and SiOH (958 cm−1) were observed. The stretching vibrations of C-H occurred at 2792 cm−1 of TEOS, 2956 cm−1 of PDMS, and 2966 cm−1 of SiNPs while the peaks of Si-C (1251 cm−1 and 1263 cm−1) were respectively observed in the spectra of PDMS and SiNPs [24]. Raman and FTIR were employed to examine the chemical structure of SiNPs synthesized from PDMS and TEOS. Figure 6 shows that the Raman spectrum of SiNPs exhibited strong PDMSs peaks appearing at 2909–2968 cm−<sup>1</sup> , 1409–1459 cm−<sup>1</sup> (C-H groups), and 706–782 cm−<sup>1</sup> (Si-C groups). These results were also found in the literature [35–37]. Furthermore, the peak at 483 cm−<sup>1</sup> were attributed to Si-O-Si [38]. These characteristic peaks indicated that the reaction of PDMS and TEOS through the sol-gel process was successful. Comparing the spectra of TEOS, PDMS, and SiNPs in Figure 7, the peaks of Si-O-Si (1074 cm−<sup>1</sup> , 808 cm−<sup>1</sup> , 796 cm−<sup>1</sup> , 786 cm−<sup>1</sup> , 495 cm−<sup>1</sup> , 480 cm−<sup>1</sup> , 460 cm−<sup>1</sup> ) and SiOH (958 cm−<sup>1</sup> ) were observed. The stretching vibrations of C-H occurred at 2792 cm−<sup>1</sup> of TEOS, 2956 cm−<sup>1</sup> of PDMS, and 2966 cm−<sup>1</sup> of SiNPs while the peaks of Si-C (1251 cm−<sup>1</sup> and 1263 cm−<sup>1</sup> ) were respectively observed in the spectra of PDMS and SiNPs [24]. *Polymers* **2020**, *12*, x FOR PEER REVIEW 7 of 15

**Raman Shift (cm-1)**

**Figure 6.** Raman spectrum of SiNPs. **Figure 6.** Raman spectrum of SiNPs.

**Figure 7.** FTIR spectra of TEOS, PDMS, and SiNPs.

4000 3600 3200 2800 2400 2000 1600 1200 800 400

**Wavenumber (cm-1)**

poly(HEMA-co-NVP)-SiNPs. These lenses appeared transparent.

2972

Figure 8 shows the photos of hydrated soft lenses including poly(HEMA-co-NVP) and

2966 1263

808 2956 1251

460

495

480

786 958

796

1074

*3.2. Optical Transparency* 

**Transmittance (%)**

**TEOS**

**PDMS**

**SiNPs**

**Intensity (a.u)**

483

709 782

**Figure 6.** Raman spectrum of SiNPs.

0 1000 2000 3000 4000

**Raman Shift (cm-1)**

2911

2966

1409 1459

**Figure 7.** FTIR spectra of TEOS, PDMS, and SiNPs. **Figure 7.** FTIR spectra of TEOS, PDMS, and SiNPs.

#### *3.2. Optical Transparency 3.2. Optical Transparency*

Figure 8 shows the photos of hydrated soft lenses including poly(HEMA-co-NVP) and poly(HEMA-co-NVP)-SiNPs. These lenses appeared transparent. Figure 8 shows the photos of hydrated soft lenses including poly(HEMA-co-NVP) and poly(HEMA-co-NVP)-SiNPs. These lenses appeared transparent. *Polymers* **2020**, *12*, x FOR PEER REVIEW 8 of 15

**Figure 8.** Photos of hydrated contact lenses. **Figure 8.** Photos of hydrated contact lenses.

Figure 9 shows that the light transmittance (T%) of the contact lens decreased with the increase of SiNPs content. The reduction in transparency can be attributed to the size distribution of SiNPs (see Figure 4) that caused light scattering. At a content of 1.2 wt%, the transmittance dropped to 90%. The light transmittance of a contact lens is preferred to be above 90% [39]. A higher SiNPs content Figure 9 shows that the light transmittance (T%) of the contact lens decreased with the increase of SiNPs content. The reduction in transparency can be attributed to the size distribution of SiNPs (see Figure 4) that caused light scattering. At a content of 1.2 wt%, the transmittance dropped to 90%. The light transmittance of a contact lens is preferred to be above 90% [39]. A higher SiNPs content would

would further decrease the transparency of the contact lens below 90%. Thus, in this study, the

maximum content of SiNPs was limited to 1.2 wt%.

100

0

20

40

**Light Transmittance (%)**

60

80

**Figure 9.** Light transmittance of SiNPs samples.

0 0.5 1 1.5

**SiNP content (wt%)**

HS-series HN2S-series HN5S-series

further decrease the transparency of the contact lens below 90%. Thus, in this study, the maximum content of SiNPs was limited to 1.2 wt%. would further decrease the transparency of the contact lens below 90%. Thus, in this study, the maximum content of SiNPs was limited to 1.2 wt%.

The light transmittance of a contact lens is preferred to be above 90% [39]. A higher SiNPs content

(**d**) HS12 (**e**) HN2S12 (**f**) HN5S12

**Figure 8.** Photos of hydrated contact lenses.

Figure 9 shows that the light transmittance (T%) of the contact lens decreased with the increase of SiNPs content. The reduction in transparency can be attributed to the size distribution of SiNPs

*Polymers* **2020**, *12*, x FOR PEER REVIEW 8 of 15

(**a**) HS0 (**b**) HN2S0 (**c**) HN5S0

**Figure 9.** Light transmittance of SiNPs samples.

#### **Figure 9.** Light transmittance of SiNPs samples. *3.3. Equilibrium Water Content*

Equilibrium water content (EWC) is an important index conferring comfortable wearing for patients because of the softness and wettability as well as the limitation of dry corneal eye [40,41]. In this present work, three series of soft lenses were prepared with different EWC values. The basic ingredient of these hydrogels was HEMA, which is a well-known monomer for contact lens. The other ingredient, NVP, is to increase the hydrophilicity of the hydrogel. All these formulations were crosslinked with 0.5 wt% of EGDMA.

Figure 10 shows that the presence of SiNPs in hydrogel network did not significantly affect the EWC of soft lenses. The values of EWC for the HS-, HN2-, and HN5- series varied around 34%, 42%, and 73%, respectively, regardless of the content of SiNPs. This is because that the content of SiNPs was low, and that these nanoparticles interacted little with the matrix of HEMA and NVP. In other words, a small number of nanoparticles were simply dispersed in the hydrogel matrix. On the contrary, for commercialized silicone soft lenses, hydrophobic ingredients such as TRIS, SiMA, and PDMS were incorporated into the main chains of the hydrogel and caused the reduction in EWC of contact lenses [19,22,42,43].

*3.3. Equilibrium Water Content* 

contact lenses [19,22,42,43].

crosslinked with 0.5 wt% of EGDMA.

Equilibrium water content (EWC) is an important index conferring comfortable wearing for patients because of the softness and wettability as well as the limitation of dry corneal eye [40,41]. In this present work, three series of soft lenses were prepared with different EWC values. The basic ingredient of these hydrogels was HEMA, which is a well-known monomer for contact lens. The other ingredient, NVP, is to increase the hydrophilicity of the hydrogel. All these formulations were

Figure 10 shows that the presence of SiNPs in hydrogel network did not significantly affect the EWC of soft lenses. The values of EWC for the HS-, HN2-, and HN5- series varied around 34%, 42%, and 73%, respectively, regardless of the content of SiNPs. This is because that the content of SiNPs was low, and that these nanoparticles interacted little with the matrix of HEMA and NVP. In other words, a small number of nanoparticles were simply dispersed in the hydrogel matrix. On the contrary, for commercialized silicone soft lenses, hydrophobic ingredients such as TRIS, SiMA, and

**Figure 10.** EWC of SiNPs samples. **Figure 10.** EWC of SiNPs samples.

#### *3.4. Contact Angle 3.4. Contact Angle*

Figure 11 shows that the contact angle of the soft lenses increased slightly with the content of SiNPs. This may be attributed to the inherent hydrophobic PDMS in SiNPs embedded in hydrogel matrix [24,44,45]. However, the difference was less than 5°, and all the contact angles were below 70°. Thus, the addition of SiNPs did little to change the wettability of these lenses. The slight increase may be attributed to a small quantity of hydrophobic PDMS (from SiNPs) exposed on the surface [23]. Figure 11 shows that the contact angle of the soft lenses increased slightly with the content of SiNPs. This may be attributed to the inherent hydrophobic PDMS in SiNPs embedded in hydrogel matrix [24,44,45]. However, the difference was less than 5◦ , and all the contact angles were below 70◦ . Thus, the addition of SiNPs did little to change the wettability of these lenses. The slight increase may be attributed to a small quantity of hydrophobic PDMS (from SiNPs) exposed on the surface [ *Polymers*  23]. **2020**, *12*, x FOR PEER REVIEW 10 of 15

**Figure 11.** Contact angle of SiNPs samples. **Figure 11.** Contact angle of SiNPs samples.

#### *3.5. IR Spectra of Contact Lenses 3.5. IR Spectra of Contact Lenses*

**Transmittance (%)**

Figure 12 shows that the FTIR spectra of HS-series lenses differed little as the contents of SiNPs increased from 0 wt% to 1.2 wt%. This phenomenon was observed for HN2S- and HN5S- series as well (not shown). In the spectra of HS-series lenses, peaks of OH groups appeared at 3350 cm−1 while peaks of Si-O-Si groups were detected at 1074 cm−1. Further, in the spectrum of HS-series, the presence of Si-C and C=O were respectively observed at the peaks of 1253 cm−1 and 1706 cm−1 [24]. In the spectra of HN2S- and HN5S- series, the peaks of OH (3338, 3338 cm−1) and Si-O-Si (1072, 1080 cm−1), Si-C Figure 12 shows that the FTIR spectra of HS-series lenses differed little as the contents of SiNPs increased from 0 wt% to 1.2 wt%. This phenomenon was observed for HN2S- and HN5S- series as well (not shown). In the spectra of HS-series lenses, peaks of OH groups appeared at 3350 cm−<sup>1</sup> while peaks of Si-O-Si groups were detected at 1074 cm−<sup>1</sup> . Further, in the spectrum of HS-series, the presence of Si-C and C=O were respectively observed at the peaks of 1253 cm−<sup>1</sup> and 1706 cm−<sup>1</sup> [24]. In the spectra

HS12

HS08

HS04

HS0

**Figure 12.** The FTIR spectra of HS- series SiNPs-containing contact lenses.

4000 3500 3000 2500 2000 1500 1000 500

**Wavenumber (cm-1)**

<sup>1706</sup> <sup>3350</sup>

<sup>1074</sup> <sup>1253</sup>

(1257 cm−1, 1257 cm−1) and C=O (1643 cm−1, 1631 cm−1) were also detected.

*3.5. IR Spectra of Contact Lenses* 

0

10

20

30

**Contact Angle (°)**

40

50

60

70

of HN2S- and HN5S- series, the peaks of OH (3338, 3338 cm−<sup>1</sup> ) and Si-O-Si (1072, 1080 cm−<sup>1</sup> ), Si-C (1257 cm−<sup>1</sup> , 1257 cm−<sup>1</sup> ) and C=O (1643 cm−<sup>1</sup> , 1631 cm−<sup>1</sup> ) were also detected. of HN2S- and HN5S- series, the peaks of OH (3338, 3338 cm−1) and Si-O-Si (1072, 1080 cm−1), Si-C (1257 cm−1, 1257 cm−1) and C=O (1643 cm−1, 1631 cm−1) were also detected.

peaks of Si-O-Si groups were detected at 1074 cm−1. Further, in the spectrum of HS-series, the presence of Si-C and C=O were respectively observed at the peaks of 1253 cm−1 and 1706 cm−1 [24]. In the spectra

**Figure 11.** Contact angle of SiNPs samples.

Figure 12 shows that the FTIR spectra of HS-series lenses differed little as the contents of SiNPs increased from 0 wt% to 1.2 wt%. This phenomenon was observed for HN2S- and HN5S- series as

0 0.5 1 1.5

**SiNP content (wt%)**

*Polymers* **2020**, *12*, x FOR PEER REVIEW 10 of 15

HN5S series

HN2S series

HS series

**Figure 12.** The FTIR spectra of HS- series SiNPs-containing contact lenses. **Figure 12.** The FTIR spectra of HS- series SiNPs-containing contact lenses. *Polymers* **2020**, *12*, x FOR PEER REVIEW 11 of 15

#### *3.6. Mechanical Properties 3.6. Mechanical Properties*

*3.7. Oxygen Permeability* 

HEMA polymers was affected by the SiNPs content.

Figure 13 presents the modulus and tensile strength of all HS, HN2S, and HN5S series. The moduli of all HN5S samples were not significantly affected by the SiNPs content increasing from 0 wt% to 1.2 wt%. Particularly, the moduli of all HN5S lenses were approximately 0.48 MPa for all SiNPs contents. This tendency was also found for HS and HN2S series samples. The moduli of HS and HN2S samples were approximately 0.65 MPa and 0.56 MPa, respectively. In this research, the moduli of all HN5S, HN2S, and HS series were around 0.48–0.65 MPa, which were similar to commercial lenses such as Acuvue Oasys, Acuvue Advance, and Biomedics 38 [43,46]. Hence, these SiNPs-contained contact lenses exhibited mechanical properties comparable to those commercial contact lenses. Figure 13 presents the modulus and tensile strength of all HS, HN2S, and HN5S series. The moduli of all HN5S samples were not significantly affected by the SiNPs content increasing from 0 wt% to 1.2 wt%. Particularly, the moduli of all HN5S lenses were approximately 0.48 MPa for all SiNPs contents. This tendency was also found for HS and HN2S series samples. The moduli of HS and HN2S samples were approximately 0.65 MPa and 0.56 MPa, respectively. In this research, the moduli of all HN5S, HN2S, and HS series were around 0.48–0.65 MPa, which were similar to commercial lenses such as Acuvue Oasys, Acuvue Advance, and Biomedics 38 [43,46]. Hence, these SiNPs-contained contact lenses exhibited mechanical properties comparable to those commercial contact lenses.

**Figure 13.** The modulus and strength of SiNPs samples. **Figure 13.** The modulus and strength of SiNPs samples.

hydrogels. For the HS series, the Dk increased from 9 to 29 barrer as the content of SiNPs increased to 1.2 wt%. For a more hydrophilic series, Dk of HN5 soft lenses increased rapidly from 39 to 71 barrer when the content of SiNPs increased from 0 to 1.2 wt%. Additionally, all formulations containing a higher concentration of NVP monomers exhibited higher oxygen transmissibility than others. As a result, the oxygen permeation of three formulation series including p(HEMA-*co*-NVP) and pure

Figure 14 shows that Dk is linearly depending on the SiNPs content. Furthermore, the slop

It is well-known that hydrophobic PDMS can improve oxygen transmissibility of soft lenses based on its siloxane groups (-Si(CH3)2-O-), especially silicon-oxygen bond [7,9,47]. However, the main weakness of the hydrophobic component is to impair the water absorbability of the lens, [16,24,48] thus EWC and Dk usually follow an inverse correlation. Accordingly, water is usually the limiting factor of oxygen transport for silicone hydrogel lenses [26,49]. On the contrary, for the SiNPsloaded hydrogel, the addition of SiNPs did not affect the water uptake ability while increasing the

**4. Conclusions** 

the manuscript.

#### *3.7. Oxygen Permeability*

Figure 14 shows that Dk is linearly depending on the SiNPs content. Furthermore, the slop increased with the hydrophilicity. Although the Dk increases with the EWC for conventional non-silicone hydrogels, the loading of SiNPs did accelerate the permeation of oxygen in these composite hydrogels. For the HS series, the Dk increased from 9 to 29 barrer as the content of SiNPs increased to 1.2 wt%. For a more hydrophilic series, Dk of HN5 soft lenses increased rapidly from 39 to 71 barrer when the content of SiNPs increased from 0 to 1.2 wt%. Additionally, all formulations containing a higher concentration of NVP monomers exhibited higher oxygen transmissibility than others. As a result, the oxygen permeation of three formulation series including p(HEMA-*co*-NVP) and pure HEMA polymers was affected by the SiNPs content.

It is well-known that hydrophobic PDMS can improve oxygen transmissibility of soft lenses based on its siloxane groups (-Si(CH3)2-O-), especially silicon-oxygen bond [7,9,47]. However, the main weakness of the hydrophobic component is to impair the water absorbability of the lens, [16,24,48] thus EWC and Dk usually follow an inverse correlation. Accordingly, water is usually the limiting factor of oxygen transport for silicone hydrogel lenses [26,49]. On the contrary, for the SiNPs-loaded hydrogel, the addition of SiNPs did not affect the water uptake ability while increasing the oxygen permeability of lenses, as shown in Table 1. This phenomenon broke the reverse relationship of EWC and Dk in comparison to other non SiNPs lenses. *Polymers* **2020**, *12*, x FOR PEER REVIEW 12 of 15 oxygen permeability of lenses, as shown in Table 1. This phenomenon broke the reverse relationship of EWC and Dk in comparison to other non SiNPs lenses.

For hydrogels loaded with SiNPs, the content of these nanoparticles was less than 1.2 wt% of the matrix. Although SiNPs were hydrophobic, the effect on the water absorbability was low due to this small content. Furthermore, hydrophilic matrix of the hydrogel interacted little with these hydrophobic nanoparticles. Thus, the mobility of these nanoparticles would be higher than in the case of silicone hydrogels where PDMS chains were integrated into the matrix. Apparently, higher mobility would facilitate the permeation of oxygen through the lens. Although in this study the highest Dk was 70 barrer, we believe that higher Dk could be attained for higher SiNPs content once the particle size was reduced through improving the synthesizing process of SiNPs, thereby breaking through the transparency limitation. For hydrogels loaded with SiNPs, the content of these nanoparticles was less than 1.2 wt% of the matrix. Although SiNPs were hydrophobic, the effect on the water absorbability was low due to this small content. Furthermore, hydrophilic matrix of the hydrogel interacted little with these hydrophobic nanoparticles. Thus, the mobility of these nanoparticles would be higher than in the case of silicone hydrogels where PDMS chains were integrated into the matrix. Apparently, higher mobility would facilitate the permeation of oxygen through the lens. Although in this study the highest Dk was 70 barrer, we believe that higher Dk could be attained for higher SiNPs content once the particle size was reduced through improving the synthesizing process of SiNPs, thereby breaking through the transparency limitation.

**Figure 14.** The effect of SiNPs content on Dk. **Figure 14.** The effect of SiNPs content on Dk.

from HEMA and NVP. The resultant SiNPs-loaded hydrogel lenses exhibited an unusual correlation between the oxygen permeability (Dk) and the equilibrium water content (EWC): the Dk increased with the content of silicone nanoparticles while the EWC changed insignificantly. Moreover, based on the result of the contact angle and Young's modulus, the loading of SiNPs slightly influenced the wetting surface and mechanical properties. The transparency was reduced to 91% when the content of SiNPs was 1.2 wt%, probably due to the light scattering from the nanoparticles. Further effort to reduce the particle size is underway in our lab. With this work, we demonstrate a novel approach to improve the oxygen permeability without impairing the hydrophilicity of soft contact lenses. These

**Author Contributions:** N.-P.-D.T. prepared experiments, as well as wrote the original draft. M.-C.Y. supervised the research project and finalized the manuscripts. All authors have read and agreed to the published version of

results would be beneficial to the development of soft contact lenses.

As demonstrated in this study, silicone nanoparticles (SiNPs) were synthesized from TEOS and
