The process typically starts with the hydrolysis of the silane compound, where it reacts with water to produce silanol groups. These silanol groups then undergo condensation reactions, forming siloxane linkages that bridge between polymer chains. Figure 3 shows curing mechanism of silane functional groups. This crosslinking imparts enhanced mechanical, thermal, and chemical properties to the cured material. Silane curing is widely employed in the production of adhesives, sealants, and coatings, offering improved durability and adhesion.
3.1 Spectroscopic characterization
The structure of PPG and IPDI as well as P-PU, ASPU1, and ASPU2 samples using FTIR and the structure of ASPU1, and ASPU2 samples using 1H-NMR was investigated. The FTIR spectra of PPG and IPDI as well as P-PU, ASPU1, and ASPU2 samples are shown in Fig. 3.
Comparing FTIR spectra of PPG, IPDI, and P-PU in Fig. 4 shows the changes in functional groups during the synthesis of urethane prepolymer. The disappearance of the peak related to the functional group O-H, the appearance of a peak at the 1720 cm− 1 (related to the C = O stretching vibration of the urethane group, red band) confirms the formation of urethane prepolymer with isocyanate end groups. The peak 2225 cm− 1 (Fig. 4-c, green band), which is the result of the vibration of the NCO functional group, indicates the formation of a prepolymer with isocyanate end groups [32]. The remaining isocyanate functional group is eligible for the reaction with the amine functional group of amino silane compounds. The amount of remaining NCO (3.09%) confirms the data on the FTIR spectrum of the P-PU sample.
The carbonyl urethane group (NH-CO-O) shows a strong peak at 1717 cm− 1 in Fig. 4a-e. The peak appearing at 1562 cm− 1 in Fig. 4c-e corresponds to the stretching vibration of the N-H group [33]. The corresponding peak of the ether bonds (C-O-C) in PPG appears in the region of 1100 cm− 1. The corresponding peak of (Si-O-C) group appears in the region of 817 cm− 1. The appearance of the peak is related to the urea carbonyl group (NH-CO-NH) at 1636 cm− 1 in Fig. 4c-e is due to the reaction of the amino group in amino silanes with the NCO group in urethane prepolymer. The complete removal of the isocyanate peak in 2255 cm− 1 is an indication of the reaction completion of the isocyanate group with the amine groups and hence, the complete conversion of urethane prepolymer to urethane prepolymer with trimethoxysilane end groups.
The 1H-NMR spectra of samples ASPU1 and ASPU2 are shown in Fig. 5.
According to Fig. 5, the hydrogens adjacent to the etheric oxygen in the chain backbone are at 3.5 ppm, the CH3 pendant group’s hydrogens of the etheric chain are at 1.2 ppm, the CH3 group corresponding to the IPDI monomer is in the below 1 ppm and other methylene groups inside the IPDI ring appeared about 1.7 ppm. The peaks observed at about 5 ppm are related to amino/amide group hydrogens. In addition, peaks related to trimethoxysilane groups appeared at 3.5 ppm and hydrogens related to the carbon attached to the nitrogen atom appeared at 3.2 ppm.
3.2 Cure time
Surface drying time (Tack-free time) and hardness (Shore A) were used to check the curing time of the samples.
The surface dry time of the samples obtained by the Tack free time test is listed in Table 1.
Table 1
The data of tack-free times at 23°C and a relative humidity of 90 ± 2%
Sample
|
Time (h)
|
ASPU1
|
3:30
|
ASPU2
|
1:00
|
The surface dry time of the ASPU1 sample at 90% relative humidity is significantly longer than the surface dry time of the ASPU2 sample. The shorter dry time in sample ASPU2 may be attributed to the presence of more alkoxysilane functional end groups in ASPU2 than ASPU1 sample. The more alkoxysilane functional end groups, the much shorter curing time and or the more crosslink density.
the curing is a time-consuming process and the sample becomes harder as the number of crosslinking points increases. The more crosslinking points and/or the smaller the molecular mass between crosslinks (Mc), the harder the sample. This upward trend continues until reaching full curing after which the hardness of samples become plateau. Figure 5 shows the variations of sample hardness during curing time at 90% RH.
It is clear from Fig. 5 that the hardness increases sharply in the beginning and then flatten after over 100 hours. However, the sample ASPU1 experiences lower hardness values than those of sample ASPU2 at all respective curing times. The hardness of sample ASPU2 reached 85 (shore A) after over 100 hours higher than sample ASPU1 with 75 (shore A). The higher hardness value in sample ASPU2 may be well attributed to the ability of the secondary type silane to form more crosslink points [10]. Note that the start time of the hardness test for the ASPU1 sample was delayed more than for the ASPU2 sample, which was proportional to the drying time of the samples (tack-free time).
3.3 Thermogravimetry and DMTA properties
Figure 6 shows the thermogravimetric analysis (TGA) and differential thermogravimetric (DTG) curves of modified polyurethane with primary and secondary type amino silanes (ASPU1 and ASPU2).
The TGA curve shows the weight loss of the sample as a function of temperature. The DTG curve shows the rate of weight loss as a function of temperature. As it can be seen in Fig. 6, the TGA curve of ASPU1 showed three stages of weight loss. The first stage was due to the loss of moisture, the second stage was due to the decomposition of the less thermally stable polymer segments (Segments that do not have pendent silane groups), and the third stage was due to the decomposition of the more thermally stable polymer segments (Segments that have pendent silane groups). The TGA curve of ASPU2 showed two stages of weight loss. The first stage was due to the loss of moisture and the second stage was due to the decomposition of the polymer. In the ASPU2 sample, we have used bis (3-aminopropyltrimethoxysilane) for modification and this compound has more silane functional groups compared to 3-aminopropyltrimethoxysilane (2:1), it is more likely that all segments of the polymer have silane groups. It means all segments in ASPU2 are thermally stable to the same extent.
The DTG curves showed that the maximum rate of weight loss occurred at around 400°C for both ASPU1 and ASPU2. The initial temperature of destruction (Tonset) in both samples is almost the same and is at around 350°C. Both samples show almost the same degradation behavior up to the temperature of 400°C at which the weight loss is 20%; T20%. In comparison with sample ASPU2, a decrease of about 10°C in T50% is observed for the sample ASPU1 with T50% of 460°C. The char yield for ASPU1 and ASPU2 was 10% and 13%, respectively, at 600°C. The presence of silicon in the structure causes thermal stability, and because bis (3-aminopropyltrimethoxysilane) has twice the number of silane groups compared to 3-aminopropyltrimethoxysilane, therefore the thermal stability of ASPU2, which has a silane agent of the secondary type of bis (3-aminopropyltrimethoxysilane), is higher than the thermal stability of ASPU1, which has a silane agent of the primary type of 3-aminopropyltrimethoxysilane.[11]
Figure 7 shows loss factor curve (Tan δ) and storage modulus curves for the samples ASPU1 and ASPU2.
Note that the corresponding parameters extracted from DMTA were tabulated in Table 2.
Table 2
The corresponding parameters extracted from DMTA for ASPU1 and ASPU2 samples
Sample
|
Tan δ peak
|
rubbery region
|
Tg
(°C)
|
Height
|
FWHM
(°C)
|
Area
(°C)
|
T/E’
(°C/MPa)
|
Mc
(g. mol− 1)
|
ASPU1
|
-32
|
0.42
|
29.41
|
17.74
|
27
|
241
|
ASPU2
|
-36
|
0.37
|
30.81
|
18.11
|
17
|
151
|
In the realm of one-part polyurethane-modified amino silane sealants, dynamic mechanical analysis reveals that ASPU1, characterized by a primary type amino silane, is less stiff than ASPU2 with a secondary type amino silane. As shown in Fig. 8, ASPU1 exhibiting a slightly lower storage modulus, its higher molecular weight between crosslinks (Mc) at 241 g/mol indicates a less tightly interconnected network, contributing to increased flexibility in ASPU1 compared to ASPU2. This difference in stiffness is further accentuated when considering the glass transition temperature (Tg). The Tg of ASPU1 at -32°C suggests a point at which the material transitions from a more flexible to a more rigid state. Contrastingly, ASPU2, featuring a slightly lower Tg of -36°C and the incorporation of a secondary amino silane, indicates a shift to a more rigid state at a lower temperature. While ASPU2 has a higher crosslink density, the Tg is influenced not only by crosslink density but also by molecular interactions and overall polymer structure. The presence of secondary amino silane in ASPU2 introduces specific molecular arrangements that impact the Tg differently than the primary amino silane in ASPU1. Molecular interactions and structural variations collectively contribute to the observed Tg differences, providing valuable insights into the intricate factors influencing the mechanical behavior of these sealants.
3.4 Mechanical properties
The tensile stress-strain curves of ASPU1 and ASPU2 are shown in Fig. 9. The corresponding parameters in tensile test were tabulated in Table 3.
Table 3
extracted data from Fig. 8
Sample code
|
Tensile strength (MPa)
|
Tensile modulus (MPa)
|
Toughness (J/m2)
|
Elongation at break (%)
|
ASPU1
|
2.09 ± 0.53
|
8.16 ± 1.21
|
17.00 ± 3.06
|
38.06 ± 0.15
|
ASPU2
|
2.51 ± 0.48
|
8.77 ± 0.93
|
13.63 ± 3.46
|
32.90 ± 1.92
|
As it can be seen from Fig. 8, the tensile strength and modulus of the ASPU2 sample have increased compared to the ASPU1 sample similar to the modulus changes in DMTA. The use of a secondary type of amino silane agent in the ASPU2 sample creates more crosslinks with a higher density in the sample, so it resists the applied stress more than the ASPU1 sample. The elongation at break and the area under the stress-strain curve in the ASPU2 sample have decreased compared to the ASPU1 sample, and the ASPU2 sample has become more brittle, which can be attributed to the greater number of crosslinks in the ASPU2 sample because it reduces the chain freedom, elongation at break as well as the area under the stress-strain curve (toughness) [34].
Figure 10 The results of lap shear strength for the joints prepared from ASPU1 and ASPU2 on Aluminum and PMMA substrates
In examining the lap shear strength results, the consideration of cross-link density becomes particularly insightful. Despite ASPU2 exhibiting a higher cross-link density than Aspu1, the lap shear strength (shown in Fig. 10) of ASPU1 surpassed that of ASPU2 on both aluminum and PMMA sheet substrates. This intriguing observation prompts a deeper exploration of the interplay between various adhesive properties.
One notable factor contributing to this discrepancy is the inherent difference in surface energies of aluminum and PMMA. The aluminum substrate, with a surface energy of 0.868 N. m− 1, surpasses the significantly lower surface energy of PMMA at 0.041 N. m− 1. This substantial contrast in surface energies could play a pivotal role in dictating the adhesion behavior of the amino silane sealants. As ASPU1 demonstrates superior lap shear strength on both substrates, it suggests that its adhesive properties, such as compatibility with higher surface energy materials, may contribute to its enhanced bonding performance.
Moreover, while ASPU2 boasts a higher cross-link density, the effectiveness of an adhesive is not solely determined by this factor. Adhesion involves a complex interplay of chemical interactions, surface energies, and interfacial compatibility. The nuanced nature of these interactions may result in the observed performance trend, wherein the lap shear strength of ASPU2 lags behind despite its higher cross-link density. Future investigations should delve into the intricate dynamics of adhesive-substrate interactions to unravel the specific mechanisms influencing the bonding behavior of ASPU1 and ASPU2.
The failure mode was analyzed after the lap shear adhesion test. The respective fractured surfaces of the samples were captured by camera and are shown in Fig. 10.
Upon scrutinizing post lap shear test photographs (as shown in Fig. 11), distinctive failure modes emerged for ASPU1 and ASPU2, shedding light on their diverse adhesive behaviors. ASPU1 exhibited a cohesive failure, indicative of internal sealant cohesion being compromised. In contrast, the failure of ASPU2 demonstrated an adhesive nature, suggesting detachment at the sealant-substrate interface. Notably, ASPU2, characterized by a higher cross-link density as a secondary type amino silane modified polyurethane, showcased this adhesive failure. In ASPU1 sample, the internal forces in bulk are not strong enough to prevent the crack propagation through the adhesive layer. Whereas in ASPU2 sample, the higher crosslink density favors the adhesive modulus, hence result in crack propagation mostly not in adhesive layer but through interfacial region. That is why the adhesive mode is the dominant failure mode in sample ASPU2. As the modulus of the adhesive layer increases (as in ASPU2 sample), the possibility of crack growth through the interfacial region (adhesive failure mode) increases. The juxtaposition of cohesive failure in ASPU1 with the adhesive failure in ASPU2 underscores the intricate interplay between cross-link density, adhesive properties, and substrate interactions, offering crucial insights into the distinct behaviors of these amino silane sealants in lap shear applications.