Hot-melt pressure-sensitive adhesive for seamless bonding of nylon fabric Part II: Process parameter optimization for seamless bonding of nylon fabric

Abstract

This study develops hot melt pressure sensitive adhesives (HMPSAs) for the seamless bonding of nylon fabric, using butyl acrylate as the main monomer material and mixing the functional monomer for polymerization. It is combined with 2-10phr diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide for the photoinitiator and ultraviolet irradiation is used to make a pre-polymer. The effects of butyl acrylate content, type of functional monomer, and 2-10phr diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide content on the molecular weight of acrylate pre-polymer are discussed, following the Taguchi method. The pre-polymer is then mixed with the reactive diluent glycidyl methacrylate blend and with 2-10phr diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide, coated on a release film, irradiated by ultraviolet light, and cured into hot melt pressure sensitive adhesives. The adhesive properties of hot melt pressure sensitive adhesive bonding on nylon include the peel strength, the shear strength, adhesive warpage, adhesive color difference, and adhesive overflow, which are discussed following the Taguchi method and the elimination and choice translating reality method for multi-quality analysis. Hot melt pressure sensitive adhesives are implemented by optimization parameters for practical validation. The results show that the peel strength of hot melt pressure sensitive adhesives is 1.495 kg/cm, the shear strength of hot melt pressure sensitive adhesives is 14.326 kg/cm², adhesive warpage is 0.93 mm, adhesive color difference is 1.66, and adhesive overflow is 0.97 mm. The performance of HMPSAs in this study is enhanced effective.

Keywords

butyl acrylate main monomer functional monomer hot melt pressure sensitive adhesives seamless bonding photoinitiator

Hot melt pressure sensitive adhesives (HMPSAs) are prepared by bulk or solution polymerization. The auto acceleration and gel effects of bulk polymerization generate heat, which is difficult to release rapidly and initiates an implosion. The polymer generates a gel and the partial products cannot be used. In solution polymerization, an organic solvent is added to the condensing system. The recirculating flow releases the heat of the free radical reaction to solve the issue in bulk polymerization.¹ However, this is unlikely to recover the organic solvent in the preparation process, leading to volatile organic compound pollution.² Therefore, this study employs an ultraviolet (UV) curing technique to avoid the implosion of bulk polymerization and the organic solvent issue of solution polymerization.

This paper aims to rigorously develop the polymerization of HMPSAs, which are applied in nylon fabric adhesion. In part I,³ a series of HMPSAs from butyl acrylate (BA), with acrylic acid (AA), beta-carboxyethyl acrylate (CEA), and hydroxyethyl acrylate (HEA) as the functional monomer, were synthesized. It was observed that AA is the best functional monomer.

The processing parameters of HMPSA polymerization are complicated. The traditional experimental methods include trial and error, one factor at a time, fractional-factorial experiments, and full-factorial experiments,⁴ which need to be improved through complex experiments, systematization, and reproducibility. Nazari et al.⁵ used the Taguchi method to select control factors, which were substituted in the orthogonal arrays for the experiment. The factors that influenced the compressive strength of aluminosilicate polymers were then illustrated. Kuo et al.⁶ applied the Taguchi method to the process parameters of a sueding machine to analyze the interaction between parameters for maximum surface softness. Ashassi-Sorkhabi et al.⁷ employed the Taguchi method as the experimental design to optimize the factors on the synthesis of a polypyrrole-Au nanocomposite coating. Kuo et al.⁸ presented the optimum setting to give uniform spinline tension using experiment plans in the Taguchi method and identified the significant process parameters. From this aforementioned literature, the Taguchi method can help reduce costs and increase the quality of products.

To verify the results of part I and optimize the performance of HMPSAs, the Taguchi method is used in this study to design and optimize the process parameters for polymerization in part II. After the Taguchi method is applied to the optimization of a single quality, the effects of multiple qualities are then attained by a multi-quality optimization method. Therefore, the elimination and choice translating reality method (ELECTRE) is applied for multi-quality analysis, which is conducted to find out the optimization parameter combination.

Free-radical polymerization theory

In this study, the main polymerization principle is free radical polymerization,⁹ including the synthesis of pre-polymers and the preparation of HMPSA polymers. First, the monomer BA and functional monomer are mixed with 2-10phr diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide (TPO) and irradiated by UV light for curing. The free radical reaction mechanism comprises three steps: initiation, chain growth, and chain termination.
The initial stage

After the photoinitiator addition and UV light irradiation, the photoinitiator is pyrolyzed to generate free radicals. The free radicals combine with the double bond of acrylate monomer and produce more free radicals, as shown in Figure 1.
Chain growth
Figure 1.
The initial reaction of the methyl methacrylate.

The initiated monomer generates new free radicals that are associated with the double bonds on other acrylates continuously. The molecular weight increases and a long molecular chain is formed, as shown in Figure 2.
Chain termination
Figure 2.
The chain growth reaction.

When the monomer is consumed, the collision between two free radical molecular chains will formulate the chemical bond. When all the free radical molecular chains have formed a chemical bond, the polymerization is terminated, as shown in Figure 3.
Figure 3.
The chain termination reaction.

Czech et al.¹⁰ presented the cationic UV-crosslinking of an acrylic PSA containing oxirane groups in the structure and with the cationic photoinitiator 1,10-bis[N,N′-(2-methylbenzothiazolium)] decane diodide to promote properties such as the tack, peel adhesion, and shear strength of self-adhesive polymer layers. Wu et al.¹¹ grafted the photoinitiator 4-maleimidobenzophenone on styrenic block copolymers as the main constituent and cured it with UV light to make HMPSAs. The results showed that the photoinitiator monomer had been grafted on polystyrene-block-polybutadiene-block-polystyrene, and the grafting rate could be controlled by the initiator concentration of benzoyl peroxide. The physical characteristics conformed to the range of the biomedical application.

Experiment

For the objective of softening temperature (70–100℃) of the HMPSAs, the soft monomer BA is selected as the main monomer of HMPSAs. BA is combined with the functional monomers AA, CEA, and HEA. The glass transition temperatures (Tg) of BA, CEA, and HEA are 106℃, 37℃, and -15℃, respectively. The three functional monomers can enhance the adhesion of HMPSAs. BA and the functional monomer are blended according to different proportions, combined with TPO and glycidyl methacrylate (GMA), and then the optimum mix proportion at the softening temperature of 70–100℃ can be found. The configuration of the Taguchi method L₉ orthogonal array experiments is shown in Tables 1 and 2.
Table 1.
The design of parameters and levels for the experiment

Experiment Factors Level

1 2 3

Pre-polymer molecular weight optimization analysis A Content of BA (wt%) 85 90 95

B Functional monomer CEA AA HEA

C Content of TPO (phr) 6 8 10

HMPSA multi-quality optimization analysis A Content of GMA (wt%) 20 30 40

B Content of TPO (phr) 6 8 10

C Photocuring time (min) 3 6 9

AA: acrylic acid; BA: butyl acrylate; CEA: beta-carboxyethyl acrylate; GMA: glycidyl methacrylate; HEA: hydroxyethyl acrylate; HMPSA: hot melt pressure sensitive adhesive; TPO: 2-10phr diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide.

Table 2.
L₉ orthogonal array experiments

Exp. No. Pre-polymer molecular weights optimization analysis
HMPSA multi-quality optimization analysis

A B C A B C

Content of BA (wt%) Functional monomer Content of TPO (phr) Content of GMA (wt%) Content of TPO (phr) Photocuring time (min)

1 85 CEA 6 20 6 3

2 85 AA 8 20 8 6

3 85 HEA 10 20 10 9

4 90 CEA 8 30 6 6

5 90 AA 10 30 8 9

6 90 HEA 6 30 10 3

7 95 CEA 10 40 6 9

8 95 AA 6 40 8 3

9 95 HEA 8 40 10 6

AA: acrylic acid; BA: butyl acrylate; CEA: beta-carboxyethyl acrylate; GMA: glycidyl methacrylate; HEA: hydroxyethyl acrylate; HMPSA: hot melt pressure sensitive adhesive; TPO: 2-10phr diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide.

The mixture was coated on a poly (ethylene terephthalate) release film. The UV-curing machine was equipped with a high-pressure mercury lamp (400 W/cm, main wavelength: 365 nm). A mirror was applied as a reflector to avoid the heating effect of the mercury lamp. The UV energy was 4000–4200 mJ/cm² from an UV-integrator 151 UV radiometer. The 1.0-inch wide acrylic HMPSAs were prepared for the specimen, pasted between the nylon fabric under the pressure of 1 kg/cm² at 100C for 25 s, and tested.

ELECTRE

ELECTRE is utilized when a set of alternatives should be identified and evaluated with respect to a set of conflicting criteria by reflecting decision makers’ preferences.^12-14 This study uses ELECTRE for multi-quality optimization in the HMPSA fabrication process. The optimization parameters for a single quality taken from the Taguchi method are substituted in ELECTRE as the base. The peel strength of HMPSAs, the shear strength of HMPSAs, the adhesive warpage, the adhesive color difference, and the adhesive overflow are selected as the qualities for experiments, so the optimization parameters can be obtained.

ELECTRE compares the performance of different process parameters. The worst process parameters are eliminated, and multiple advantageous process parameters remain for multiple attribute decision-making analysis. The advantages of ELECTRE include straightforward decision-making logic, systematic process, and clarifying the performance of various proposals.

The major decision-making steps for ELECTRE are as follows:
Establishment of the normalized decision matrix

The evaluation of m process parameters under n attribute criteria of qualities is represented in a matrix, expressed as $P = [p_{ij}]_{m \times n}$ . The P matrix is normalized, expressed as $R = [R_{ij}]_{m \times n}$ , and shown as:
$\begin{matrix} R_{ij} = \frac{p_{ij}}{\sqrt{\sum_{i = 1}^{m} p_{ij}^{2}}} \end{matrix}$
(1)
where m is the number of experiments, and n is the number of the quality.
Calculation of the weighted normalized decision matrix V

The weights of various qualities result in a weighting matrix:
$W_{j} = [\begin{matrix} W_{1} & \dots & W_{n} \end{matrix}]$
(2)
and
$\begin{matrix} \sum_{i = n}^{n} W_{i} = 1 \end{matrix}$
(3)
where W_j is the weight of each quality.

The normalized matrix (R) is multiplied by the weighting matrix (W) to obtain the weighted normalized matrix (V).
$\begin{matrix} V = RW = [\begin{matrix} v_{11} & v_{12} & \dots & v_{1 n} \\ v_{21} & v_{22} & \dots & v_{2 n} \\ : & : & : \\ v_{m 1} & v_{m 2} & \dots & v_{mn} \end{matrix}] \end{matrix}$
(4)

Determination of the concordance set and discordance set

In the weighted normalized matrix V, any two process parameters are compared. In it, the i₁ parameter of the quality κ is defined as $v_{i_{1} κ}$ and the i₂ parameter of κ as $v_{i_{2} κ}$ . If $v_{i_{1} κ} > v_{i_{2} κ}$ , then the element κ is classified as the concordance matrix set $c_{ij}$ . On the contrary, if $v_{i_{1} κ} \leq v_{i_{2} κ}$ , then the element κ is classified as the discordance matrix set $d_{ij}$ .
$\begin{matrix} c_{ij} = {κ | v_{i_{1} κ} > v_{i_{2} κ}} \end{matrix}$
(5)

$\begin{matrix} d_{ij} = {κ | v_{i_{1} κ} \leq v_{i_{2} κ}} \end{matrix}$
(6)
where $κ = 1, 2, \dots, n$
Set up the concordance matrix (C-Matrix)

When the process parameter $p_{i_{1} j}$ is better than $p_{i_{2} j}$ , it is classified as the concordance matrix set, $c_{ij}$ . The weights of attribute quality represented by various elements in each concordance set are added up. The sum forms the concordance index $c'_{ij}$ .
$c'_{ij} = \frac{\sum_{k \in c_{ij}} w_{k}}{\sum_{k = 1}^{N} w_{k}}$
(7)
where the sum of all attribute weights is 1.

Here, $c'_{ij}$ can be reduced as:
$c'_{ij} = \sum_{k \in c_{ij}} w_{k}, 0 \leq c'_{ij} \leq 1$
(8)

The concordance index $c'_{ij}$ forms the C-Matrix.
$\begin{matrix} C = [\begin{matrix} c'_{11} & c'_{12} & \dots & \dots & c'_{1 m} \\ c'_{21} & c'_{22} & \dots & \dots & c'_{2 m} \\ : & : & : \\ : & : & : \\ c'_{m 1} & c'_{m 2} & \dots & \dots & c'_{mm} \end{matrix}] \end{matrix}$
(9)

Set up the discordance matrix (D-Matrix)

When the process parameter $p_{i_{1} j}$ is worse than $p_{i_{2} j}$ , it is classified as the discordance matrix and set as $d_{ij}$ . This D-Matrix $d'_{ij}$ is shown as:
$\begin{matrix} d'_{ij} = \frac{{max}_{k \in D_{ij}} | v_{i_{1} k} - v_{i_{2} k} |}{{max}_{k} | v_{i_{1} k} - v_{i_{2} k} |} \end{matrix}$
(10)

The discordance index $d'_{ij}$ forms the discordance matrix, D-Matrix, $D = | d'_{ij} |_{M \times M}$
$\begin{matrix} D = [\begin{matrix} d'_{11} & d'_{12} & \dots & \dots & d'_{1 m} \\ d'_{21} & d'_{22} & \dots & \dots & d'_{2 m} \\ : & : & : \\ : & : & : \\ d'_{m 1} & d'_{m 2} & \dots & \dots & d'_{mm} \end{matrix}] \end{matrix}$
(11)

Determination of the concordance dominance matrix (F-Matrix)

The concordance threshold value ( $\bar{c}$ ) is now determined. Here, $\bar{c}$ represents the average of all elements in the concordance matrix.
$\bar{c} = \frac{\sum_{\binom{i = 1}{i \neq j}}^{m} \sum_{\binom{j = 1}{j \neq i}}^{m} c'_{ij}}{m (m - 1)}$
(12)

According to $\bar{c}$ , the concordance matrix, F-Matrix, is obtained.
$\begin{matrix} F = | f_{ij} |_{m \times m} \\ {\begin{matrix} f_{ij} = 1, if c'_{ij} > \bar{c} \\ f_{ij} = 0, if c'_{ij} \leq \bar{c} \end{matrix} \end{matrix}$
(13)

Determination of the discordance dominance matrix (G-Matrix)

The discordance threshold value ( $\bar{d}$ ) is determined. Here, $\bar{d}$ represents the average of all elements in the discordance matrix:
$\bar{d} = \frac{\sum_{\binom{i = 1}{i \neq j}}^{m} \sum_{\binom{j = 1}{j \neq i}}^{m} d'_{ij}}{m (m - 1)}$
(14)

The main polymerization principle; according to $\bar{d}$ , the G-Matrix is obtained.
$\begin{matrix} G = | g_{ij} |_{m \times m} \\ {\begin{matrix} g_{ij} = 1, if d'_{ij} < \bar{d} \\ g_{ij} = 0, if d'_{ij} \geq \bar{d} \end{matrix} \end{matrix}$
(15)

Determination of the aggregate dominance matrix (E-Matrix)

The products of the F-Matrix and the G-Matrix are set up, resulting in the aggregate dominance E-Matrix.
$E = | e_{ij} |_{m x m}, e_{ij} = f_{ij} \times g_{ij}$
(16)

Establishment of the less-favorable alternatives

From the E-Matrix, $e_{ij} = 1$ indicates that the vertical process parameter is better than the horizontal process parameter, so the horizontal process parameter can be removed. Here, $e_{ij} = 0$ indicates that the vertical process parameter is worse than or identical to the horizontal process parameter, so that the vertical process parameter can be removed.

Results and discussion

There are two experiments for single-quality optimization. The first is the synthesis of the pre-polymer. BA is used as the main monomer, mixed with three functional monomers, CEA, AA, and HEA, then mixed with TPO for UV polymerization. The second experiment examines the adhesion properties of HMPSAs on nylon fabric. The pre-polymer of optimum parameters in the first experiment is mixed with different proportions of GMA and TPO and stuck on a nylon fabric.

Optimization analysis of pre-polymer molecular weights

The property of the molecular weight of the pre-polymer is tested by the standard of ASTM E386-90, whereby the larger it is, the better. The signal-to-noise (S/N) ratio¹⁵ is calculated by the experiment results and shown in Table 3.
Table 3.
S/N ratio of the molecular weight of BA pre-polymer

Molecular weight Exp. No. y₁ y₂ y₃ Average S/N ratio (dB)

1 48,387 48,863 49,981 49,077.00 93.82

2 112,891 113,838 106,964 111,231.00 100.91

3 82,401 85,189 83,935 83,841.67 98.47

4 47,720 46,539 47,102 47,120.33 93.46

5 117,162 116,496 125,894 119,850.67 101.56

6 91,305 92,085 91,874 91,754.67 99.25

7 36,702 34,282 35,468 35,484.00 90.99

8 103,127 101,762 99,794 101,561.00 100.13

9 99,385 98,892 99,351 99,209.33 99.93

*
y₁ to y₃ denotes the observations of the BA pre-polymer molecular weight.

BA: butyl acrylate; S/N: signal to noise.

The molecular weights of the various experiments correspond to the levels of the orthogonal array. The response value of each factor at various levels can be calculated by main effect analysis¹⁵ as shown in Table 4 and Figure 4.
Table 4.
Response table of the molecular weight of BA pre-polymer (unit: dB)

Factors Level Content of BA Functional monomer Content of TPO

Level 1 97.73 92.76 97.73

Level 2 98.09 100.87 98.10

Level 3 97.02 99.22 97.00

Maximum response 98.09 100.87 98.10

Minimum response 97.02 92.76 97.00

Effect of factor 1.07 8.11 1.10

BA: butyl acrylate; TPO: 2-10phr diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide.

Figure 4.
Response figure of the molecular weight of BA pre-polymer.

According to Table 4 and Figure 4, the optimum factors are A₂, B₂ and C₂, representing a BA monomer proportion of 90 wt%, functional monomer AA, and a TPO content of 8 phr. AA is shown as the optimal functional monomer of molecular weight, and the molecular weight decreased when the content of BA exceeded 90% and the content of TPO exceeded 10 phr. The most influential factor on the molecular weight is the functional monomer, followed by TPO and BA content.

HMPSA single-quality optimization analysis

According to the single-quality optimization of the BA pre-polymer, a BA monomer ratio of 90 wt%, functional monomer AA and TPO content pf 8 phr are the optimum conditions for the pre-polymer. The maximum average molecular weight is 116,175.33 according to the confirmation experiment. Therefore, this study applies this group of optimization pre-polymers to fabricate the HMPSAs. The three control factors of GMA content, TPO content and irradiation time are selected. Three levels are substituted in the Taguchi L₉ orthogonal array for a single-quality optimization experiment. The main effect analysis is implemented for the properties, such as the peel strength of HMPSAs, the shear strength of HMPSAs, the adhesive warpage, the adhesive color difference, and the adhesive overflow.
Peel strength optimization result

The property of the shear strength of HMPSAs is tested by the standard of ASTM D3330, and the larger it is, the better. The S/N ratio is calculated by experiment results and shown in Table 5.
Table 5.
Analysis results of the peel strength of HMPSAs

Peel strength Exp. No. y₁ (kg/cm) y₂ (kg/cm) y₃ (kg/cm) Average S/N ratio (dB)

1 582 573 594 583.00 55.31

2 993 1038 974 1001.67 60.01

3 417 398 465 426.67 52.55

4 1062 1134 997 1064.33 60.51

5 1497 1472 1506 1064.33 60.51

6 1094 1124 998 583.00 55.31

7 636 628 647 637.00 56.08

8 824 819 832 825.00 58.33

9 563 551 574 562.67 57.22

*
y₁ to y₃ denotes the observations of the peel strength of HMPSAs.

HMPSA: hot melt pressure sensitive adhesive; S/N: signal to noise.

The response table and figure of the peel strength of HMPSAs are shown in Table 6 and Figure 5.
Table 6.
Response table of the peel strength of HMPSAs (unit: dB)

Factors Level Content of GMA Content of TPO Photocuring time

Level 1 55.95 57.30 56.32

Level 2 58.77 59.61 59.24

Level 3 57.21 55.02 56.38

Maximum response 58.77 59.61 59.24

Minimum response 55.95 55.02 56.32

Effect of factor 2.82 4.59 2.92

HMPSA: hot melt pressure sensitive adhesive; TPO: 2-10phr diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide.

Figure 5.
Response figure of the peel strength of hot melt pressure sensitive adhesives (HMPSAs).

According to Table 6 and Figure 5, the optimum factors are A₂, B_2, and C₂, representing the GMA content of 30 wt%, TPO content of 8 phr, and an irradiation time of 6 min. The peel strength of HMPSAs decreased when the content of GMA exceeded 30%, the content of TPO exceeded 8 phr, and the photocuring time exceeded 6 min. The most influential factor on the peel strength of HMPSAs is the content of TPO, followed by the photocuring time and the content of GMA.
Shear strength optimization result

The property of the shear strength of HMPSAs is tested by the standard of ASTM D3330, and the larger it is, the better. The S/N ratio is calculated by experiment results and shown in Table 7.
Table 7.
Analysis results of the shear strength of HMPSAs

Shear strength Exp. No. y₁ (kg/cm²) y₂ (kg/cm²) y₃ (kg/cm²) Average S/N ratio (dB)

1 9169 8976 9684 9276.33 79.33

2 12,484 12,193 12,698 12,458.33 81.91

3 7322 7068 7682 7357.33 77.32

4 11,861 11,237 12,069 11,722.33 81.37

5 13,527 13,294 14,164 13,661.67 82.70

6 13,097 12,946 13,182 13,075.00 82.33

7 12,231 11,994 12,681 12,302.00 81.79

8 12,248 12,582 11,967 12,265.67 81.77

9 12,136 12,487 11,853 12,158.67 81.69

*
y₁ to y₃ denotes the observations of the shear strength of HMPSAs.

HMPSA: hot melt pressure sensitive adhesive; S/N: signal to noise.

The response table and figure of the shear strength of HMPSAs are shown in Table 8 and Figure 6.
Table 8.
Response table of the shear strength of HMPSAs (unit: dB)

Factors Level Content of GMA Content of TPO Photocuring time

Level 1 79.52 80.83 81.14

Level 2 82.13 82.12 81.66

Level 3 81.75 80.45 80.60

Maximum response 82.13 82.12 81.66

Minimum response 79.52 80.45 80.60

Effect of factor 2.61 1.67 1.06

HMPSA: hot melt pressure sensitive adhesive; TPO: 2-10phr diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide.

Figure 6.
Response figure of the shear strength of hot melt pressure sensitive adhesives (HMPSAs).

According to Table 8 and Figure 6, the optimum factors are A₂, B_2, and C₂, representing the GMA content of 30 wt%, TPO content of 8 phr, and an irradiation time of 6 min. The shear strength of HMPSAs decreased when the content of GMA exceeded 30%, the content of TPO exceeded 8 phr, and the photocuring time exceeded 6 min. The most influential factor on the shear strength of HMPSAs is the content of GMA, followed by the content of TPO and the photocuring time.
Adhesive warpage optimization result

A smaller property of the adhesive warpage is better. The S/N ratio is calculated from the experimental results and shown in Table 9.
Table 9.
Analysis results of the adhesive warpage

Warpage Exp. No. y₁ (mm) y₂ (mm) y₃ (mm) Average S/N ratio (dB)

1 0.8 0.9 0.7 0.80 1.89

2 1.0 1.1 1.2 1.10 − 0.85

3 1.1 1.0 0.9 1.00 − 0.03

4 0.9 1.0 0.9 0.93 0.59

5 1.0 1.0 1.1 1.03 − 0.29

6 1.0 1.0 1.2 1.07 − 0.59

7 0.9 1.0 1.1 1.00 − 0.03

8 0.9 1.1 1.1 1.03 − 0.32

9 1.0 1.2 1.0 1.07 − 0.59

*
y₁ to y₃ denotes the observations of adhesive warpage.

S/N: signal to noise.

The response table and figure of the adhesive warpage are shown in Table 10 and Figure 7.
Table 10.
Response table of the adhesive warpage (unit: dB)

Factor Level Content of GMA Content of TPO Photocuring time

Level 1 0.34 0.82 0.33

Level 2 −0.10 −0.49 −0.29

Level 3 −0.31 −0.41 −0.12

Maximum response 0.34 0.82 0.33

Minimum response −0.31 −0.49 −0.29

Effect of factor 0.65 1.31 0.62

TPO: 2-10phr diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide.

Figure 7.
Response figure of the adhesive warpage.

According to Table 10 and Figure 7, the optimum factors are A₁, B₁, and C₁, representing the GMA content of 20 wt%, TPO content of 6 phr, and an irradiation time of 3 min. The adhesive warpage decreased when the content of GMA, the content of TPO and the photocuring increased. The most influential factor on the adhesive warpage of HMPSAs is the content of TPO, followed by the content of GMA and the photocuring time.
Adhesive color difference optimization result

The property of the adhesive color difference is tested by the standard of ASTM E308-15, and the smaller it is, the better. The S/N ratio is calculated from the experimental results and shown in Table 11.
Table 11.
Analysis results of the adhesive color difference

Color difference Exp. No. y₁ y₂ y₃ Average S/N ratio (dB)

1 1.71 1.82 1.69 1.74 −4.82

2 1.83 1.84 1.81 1.83 −5.23

3 1.86 1.85 1.87 1.86 −5.39

4 1.73 1.75 1.7 1.73 −4.74

5 1.83 1.8 1.85 1.83 −5.23

6 1.89 1.92 1.86 1.89 −5.53

7 1.79 1.81 1.76 1.79 −5.04

8 1.83 1.86 1.8 1.83 −5.25

9 1.9 1.92 1.88 1.90 −5.58

*
y₁ to y₃ denotes the observations of the adhesive color difference.

S/N: signal to noise.

The response table and figure of the adhesive color difference are shown in Table 12 and Figure 8.
Table 12.
Response table of the adhesive color difference (unit: dB)

Factors Level Content of GMA Content of TPO Photocuring time

Level 1 −5.15 −4.87 −5.20

Level 2 −5.17 −5.24 −5.18

Level 3 −5.29 −5.50 −5.22

Maximum response −5.15 −4.87 −5.18

Minimum response −5.29 −5.50 −5.22

Effect of factor 0.14 0.63 0.04

TPO: 2-10phr diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide.

Figure 8.
Response figure of the adhesive color difference.

According to Table 12 and Figure 8, the optimum factors are A₁, B₁, and C₂, representing the GMA content of 20 wt%, a TPO content of 6 phr, and an irradiation time of 6 min. The adhesive color difference decreased when the content of GMA and the content of TPO increased, and the optimal adhesive color difference obtained when the photocuring is 6 min. The most influential factor on the adhesive color difference of HMPSAs is the content of TPO, followed by the content of GMA and the photocuring time.
Adhesive overflow optimization result

A smaller adhesive overflow is better. The S/N ratio is calculated by experiment results and shown in Table 13.
Table 13.
Analysis results of the adhesive overflow

Adhesive overflow Exp. No. y₁ (mm) y₂ (mm) y₃ (mm) Average S/N ratio (dB)

1 1.0 0.9 1.2 1.03 −0.35

2 1.3 1.4 1.2 1.30 −2.30

3 1.2 1.3 1.0 1.17 −1.39

4 0.8 1.0 1.1 0.97 0.22

5 1.1 1.0 1.2 1.10 −0.85

6 1.4 1.2 1.5 1.37 −2.75

7 0.9 1.0 1.1 1.00 −0.03

8 1.1 1.0 1.3 1.13 −1.14

9 1.3 1.2 1.1 1.20 −1.60

*
y₁ to y₃ denotes the observations of the adhesive overflow.

S/N: signal to noise.

The response table and figure for the adhesive overflow are shown in Table 14 and Figure 9.
Table 14.
Response table of the adhesive overflow (unit: dB)

Factors Level Content of GMA Content of TPO Photocuring time

Level 1 −1.34 −0.05 −1.41

Level 2 −1.13 −1.43 −1.23

Level 3 −0.92 −1.91 −0.76

Maximum response −0.92 −0.05 −0.76

Minimum response −1.34 −1.91 −1.41

Effect of factor 0.42 1.86 0.65

TPO: 2-10phr diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide.

Figure 9.
Response figure of the adhesive overflow of hot melt pressure sensitive adhesives (HMPSAs).

According to Table 14 and Figure 9, the optimum factors are A₃, B₁, and C₃, representing the GMA content of 40 wt%, a TPO content of 6 phr, and an irradiation time of 9 min. The adhesive overflow decreased when the content of TPO increased. The adhesive overflow also increased when the content of GMA and the photocuring increased. The most influential factor on the adhesive overflow is the content of TPO, followed by the photocuring time and the content of GMA.

HMPSA multi-quality optimization analysis

The Taguchi method is a single-quality analysis and is used to process one quality at a time. Because this study analyzes multiple qualities, the multi-quality method is used to reveal the optimization parameters. ELECTRE is selected for multi-quality analysis; there are five qualities. For HMPSAs, a larger peel and shear strength is better; however, a smaller adhesive warpage, adhesive color difference, and adhesive overflow are better. Using ELECTRE, the aggregate dominance set matrix can be calculated as in Table 15.
Table 15.
Aggregate dominance matrix

No. 1 2 3 4 5 6 7 8 9 sum

1 – 0 1 0 0 0 0 0 0 1

2 1 – 1 0 0 0 1 1 1 5

3 0 0 – 0 0 0 0 0 0 0

4 1 0 1 – 0 1 1 1 1 6

5 1 1 1 1 – 1 1 1 1 8

6 1 0 1 0 0 – 0 0 0 2

7 1 0 1 0 0 0 – 0 0 2

8 1 0 1 0 0 1 1 – 1 5

9 1 0 1 0 0 1 1 0 – 4

In Table 15, the value 1 means the vertical proposal is better than the horizontal proposal, and 0 means the vertical proposal is worse than or identical to the horizontal proposal. The dominances of various proposals are added up to obtain the rightmost sum. The maximum sum is 8 and the minimum is 0. The combination of proposal 5 (aggregate dominance is 8) is the best of the nine proposals. The sum of the aggregate dominance matrix corresponds to the orthogonal array and the response values of each factor can be calculated by main effect analysis as shown in Table 16 and Figure 10.
Table 16.
Response table of multi-quality optimization

Content of GMA Content of TPO Photocuring time

Level 1 2.00 3.00 2.67

Level 2 5.33 6.00 5.00

Level 3 3.67 2.00 3.33

Maximum response 5.33 6.00 5.00

Minimum response 2.00 2.00 2.67

Effect of factor 3.33 4.00 2.33

TPO: 2-10phr diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide.

Figure 10.
Response figure of the multi-quality optimization.

According to Table 16 and Figure 10, the optimum factors are A₂, B₂, and C₂, representing a GMA content of 30 wt%, a TPO content of 8 phr, and an irradiation time of 6 min. The most influential factor on the multi-quality optimization is the content of TPO, followed by the content of GMA and the photocuring time.

Confirmation experiment

The optimization parameters of pre-polymer polymerization and HMPSA polymerization are used for the confirmation experiment. The optimum parameters of pre-polymer polymerization are AA 10 wt%, GMA 30 wt%, and TPO 8 phr. The optimum parameters of HMPSA polymerization are a GMA content of 30 wt%, a TPO content of 8 phr, and irradiation time of 6 min. The result of the confirmation experiment is shown in Table 17.
Table 17.
Comparison of general HMPSAs and the HMPSAs in this study

Quality Target General HMPSAs HMPSAs in this study Gain (%)

Peel strength of HMPSAs (kg/cm) >1.0 1.430 1.495 4.55%

Shear strength of HMPSAs (kg/cm²) >13 14.261 14.326 0.46%

Adhesive warpage (mm) <2 1.0 0.93 7.00%*

Adhesive color difference ΔE <2.0 1.86 1.66 10.75%*

Adhesive overflow (mm) <2 1.0 0.97 3.00%*

**
mean larger qualities are better, * mean smaller qualities are better.

HMPSAs: hot melt pressure sensitive adhesives.

The five qualities are obtained after the multi-quality optimization experiment. The peel strength of HMPSAs is increased by 4.55%, the shear strength of HMPSAs is increased by 0.46%, the adhesive warpage is reduced by 7%, the adhesive color difference is reduced by 10.75%, and the adhesive overflow is reduced by 3%. The HMPSAs in this study are better than general commercial HMPSA, meaning the multi-quality optimization experiment can effectively find the optimum experimental parameters.

Conclusion

This study uses UV irradiation to prepare HMPSAs for seamless bonding of nylon fiber fabric, and the Taguchi method and ELECTRE are applied for experimental optimization. The Taguchi method can reduce the number of experiments and cost, as well as effectively arrive at a single optimum parameter. This optimization parameter is used in the multi-quality experiment to find the experimental parameters’ optimizing multi-qualities of HMPSAs such as the peel strength, the shear strength of HMPSAs, the adhesive warpage, the adhesive color difference, and the adhesive overflow. ELECTRE is able to determine the multi-quality optimization parameter. The result of the confirmation experiment shows that the peel strength is increased by 4.55%, the shear strength is increased by 0.46%, the adhesive warpage is reduced by 7%, the adhesive color difference is reduced by 10.75%, and the adhesive overflow is reduced by 3%.

The performance of HMPSAs in this study is enhanced effectively and there is considerable potential for its use in commercial markets. At present, the HMPSAs prepared in this study are mainly used for gluing nylon fiber fabric, which is expected to be used in other fiber fabrics in the future.

Experiment	Factors	Level
Pre-polymer molecular weight optimization analysis	A	Content of BA (wt%)	85	90	95
B	Functional monomer	CEA	AA	HEA
C	Content of TPO (phr)	6	8	10
HMPSA multi-quality optimization analysis	A	Content of GMA (wt%)	20	30	40
B	Content of TPO (phr)	6	8	10
C	Photocuring time (min)	3	6	9

Exp. No.	Pre-polymer molecular weights optimization analysis	HMPSA multi-quality optimization analysis
1	85	CEA	6	20	6	3
2	85	AA	8	20	8	6
3	85	HEA	10	20	10	9
4	90	CEA	8	30	6	6
5	90	AA	10	30	8	9
6	90	HEA	6	30	10	3
7	95	CEA	10	40	6	9
8	95	AA	6	40	8	3
9	95	HEA	8	40	10	6

Molecular weight Exp. No.	y₁	y₂	y₃	Average	S/N ratio (dB)
1	48,387	48,863	49,981	49,077.00	93.82
2	112,891	113,838	106,964	111,231.00	100.91
3	82,401	85,189	83,935	83,841.67	98.47
4	47,720	46,539	47,102	47,120.33	93.46
5	117,162	116,496	125,894	119,850.67	101.56
6	91,305	92,085	91,874	91,754.67	99.25
7	36,702	34,282	35,468	35,484.00	90.99
8	103,127	101,762	99,794	101,561.00	100.13
9	99,385	98,892	99,351	99,209.33	99.93

Factors Level	Content of BA	Functional monomer	Content of TPO
Level 1	97.73	92.76	97.73
Level 2	98.09	100.87	98.10
Level 3	97.02	99.22	97.00
Maximum response	98.09	100.87	98.10
Minimum response	97.02	92.76	97.00
Effect of factor	1.07	8.11	1.10

Peel strength Exp. No.	y₁ (kg/cm)	y₂ (kg/cm)	y₃ (kg/cm)	Average	S/N ratio (dB)
1	582	573	594	583.00	55.31
2	993	1038	974	1001.67	60.01
3	417	398	465	426.67	52.55
4	1062	1134	997	1064.33	60.51
5	1497	1472	1506	1064.33	60.51
6	1094	1124	998	583.00	55.31
7	636	628	647	637.00	56.08
8	824	819	832	825.00	58.33
9	563	551	574	562.67	57.22

Factors Level	Content of GMA	Content of TPO	Photocuring time
Level 1	55.95	57.30	56.32
Level 2	58.77	59.61	59.24
Level 3	57.21	55.02	56.38
Maximum response	58.77	59.61	59.24
Minimum response	55.95	55.02	56.32
Effect of factor	2.82	4.59	2.92

Shear strength Exp. No.	y₁ (kg/cm²)	y₂ (kg/cm²)	y₃ (kg/cm²)	Average	S/N ratio (dB)
1	9169	8976	9684	9276.33	79.33
2	12,484	12,193	12,698	12,458.33	81.91
3	7322	7068	7682	7357.33	77.32
4	11,861	11,237	12,069	11,722.33	81.37
5	13,527	13,294	14,164	13,661.67	82.70
6	13,097	12,946	13,182	13,075.00	82.33
7	12,231	11,994	12,681	12,302.00	81.79
8	12,248	12,582	11,967	12,265.67	81.77
9	12,136	12,487	11,853	12,158.67	81.69

Factors Level	Content of GMA	Content of TPO	Photocuring time
Level 1	79.52	80.83	81.14
Level 2	82.13	82.12	81.66
Level 3	81.75	80.45	80.60
Maximum response	82.13	82.12	81.66
Minimum response	79.52	80.45	80.60
Effect of factor	2.61	1.67	1.06

Warpage Exp. No.	y₁ (mm)	y₂ (mm)	y₃ (mm)	Average	S/N ratio (dB)
1	0.8	0.9	0.7	0.80	1.89
2	1.0	1.1	1.2	1.10	− 0.85
3	1.1	1.0	0.9	1.00	− 0.03
4	0.9	1.0	0.9	0.93	0.59
5	1.0	1.0	1.1	1.03	− 0.29
6	1.0	1.0	1.2	1.07	− 0.59
7	0.9	1.0	1.1	1.00	− 0.03
8	0.9	1.1	1.1	1.03	− 0.32
9	1.0	1.2	1.0	1.07	− 0.59

Factor Level	Content of GMA	Content of TPO	Photocuring time
Level 1	0.34	0.82	0.33
Level 2	−0.10	−0.49	−0.29
Level 3	−0.31	−0.41	−0.12
Maximum response	0.34	0.82	0.33
Minimum response	−0.31	−0.49	−0.29
Effect of factor	0.65	1.31	0.62

Color difference Exp. No.	y₁	y₂	y₃	Average	S/N ratio (dB)
1	1.71	1.82	1.69	1.74	−4.82
2	1.83	1.84	1.81	1.83	−5.23
3	1.86	1.85	1.87	1.86	−5.39
4	1.73	1.75	1.7	1.73	−4.74
5	1.83	1.8	1.85	1.83	−5.23
6	1.89	1.92	1.86	1.89	−5.53
7	1.79	1.81	1.76	1.79	−5.04
8	1.83	1.86	1.8	1.83	−5.25
9	1.9	1.92	1.88	1.90	−5.58

Factors Level	Content of GMA	Content of TPO	Photocuring time
Level 1	−5.15	−4.87	−5.20
Level 2	−5.17	−5.24	−5.18
Level 3	−5.29	−5.50	−5.22
Maximum response	−5.15	−4.87	−5.18
Minimum response	−5.29	−5.50	−5.22
Effect of factor	0.14	0.63	0.04

Adhesive overflow Exp. No.	y₁ (mm)	y₂ (mm)	y₃ (mm)	Average	S/N ratio (dB)
1	1.0	0.9	1.2	1.03	−0.35
2	1.3	1.4	1.2	1.30	−2.30
3	1.2	1.3	1.0	1.17	−1.39
4	0.8	1.0	1.1	0.97	0.22
5	1.1	1.0	1.2	1.10	−0.85
6	1.4	1.2	1.5	1.37	−2.75
7	0.9	1.0	1.1	1.00	−0.03
8	1.1	1.0	1.3	1.13	−1.14
9	1.3	1.2	1.1	1.20	−1.60

Factors Level	Content of GMA	Content of TPO	Photocuring time
Level 1	−1.34	−0.05	−1.41
Level 2	−1.13	−1.43	−1.23
Level 3	−0.92	−1.91	−0.76
Maximum response	−0.92	−0.05	−0.76
Minimum response	−1.34	−1.91	−1.41
Effect of factor	0.42	1.86	0.65

No.	1	2	3	4	5	6	7	8	9	sum
1	–	0	1	0	0	0	0	0	0	1
2	1	–	1	0	0	0	1	1	1	5
3	0	0	–	0	0	0	0	0	0	0
4	1	0	1	–	0	1	1	1	1	6
5	1	1	1	1	–	1	1	1	1	8
6	1	0	1	0	0	–	0	0	0	2
7	1	0	1	0	0	0	–	0	0	2
8	1	0	1	0	0	1	1	–	1	5
9	1	0	1	0	0	1	1	0	–	4

	Content of GMA	Content of TPO	Photocuring time
Level 1	2.00	3.00	2.67
Level 2	5.33	6.00	5.00
Level 3	3.67	2.00	3.33
Maximum response	5.33	6.00	5.00
Minimum response	2.00	2.00	2.67
Effect of factor	3.33	4.00	2.33

Quality	Target	General HMPSAs	HMPSAs in this study	Gain (%)
Peel strength of HMPSAs (kg/cm)	>1.0	1.430	1.495	4.55%**
Shear strength of HMPSAs (kg/cm²)	>13	14.261	14.326	0.46%**
Adhesive warpage (mm)	<2	1.0	0.93	7.00%*
Adhesive color difference	ΔE <2.0	1.86	1.66	10.75%*
Adhesive overflow (mm)	<2	1.0	0.97	3.00%*

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research was supported by the Ministry of Science and Technology of the Republic of China under Grant No. 104-2745-8-011-004.

References

Zhang

Peng

et al.

A green approach to porous and dense antifouling membranes through solvent-free bulk polymerization. Chem Eng Sci 2015; 135: 501–508.

Czech

. Development of solvent-free pressure-sensitive adhesive acrylics. Int J Adhes Adhes 2004; 24: 119–125.

Kuo CFJ

Chen

Lin

et al.

Hot-melt pressure-sensitive adhesive for seamless bonding of nylon fabric part I: Effect of a functional monomer. Text Res J. 2019; 89: 926–935.

Kuram

Tasci

Altan

et al.

Investigating the effects of recycling number and injection parameters on the mechanical properties of glass-fibre reinforced nylon 6 using Taguchi method. Mater Des 2013; 49: 139–150.

Nazari

Khanmohammadi

Amini

et al.

Production geopolymers by Portland cement: Designing the main parameters’ effects on compressive strength by Taguchi method. Mater Des 2012; 41: 43–49.

Kuo

CFJ

Lin

. Design and verification of fabric surface softness testing system. Text Res J 2011; 81.16: 1724–1732.

Ashassi‐Sorkhabi

Bagheri

Rezaei‐Moghadam

. Protective properties of PPy-Au nanocomposite coatings prepared by sonoelectrochemisty and optimized by the Taguchi method. J Appl Polym Sci 2014; 131(22): 41087–41087.

Kuo

CFJ

Huang

Gao

et al.

Recognition of fault process conditions based on spinline tension in melt spinning. Text Res J 2014; 84(14): 1549–1557.

Savin

. Writing reaction mechanisms in organic chemistry, Cambridge: Academic Press. Third edition, 2014, pp. 23–292.

10.

Czech

Kabatc

Ragańska

et al.

2-Methylbenzothiazolium derivatives as cationic photoreactive crosslinker for acrylic pressure-sensitive adhesives containing oxirane groups from glycidyl methacrylate. Int J Adhes Adhes 2018; 80: 39–42.

11.

Jiang

et al.

A novel UV-crosslinked pressure-sensitive adhesive based on photoinitia-tor-grafted SBS. Int J Adhes Adhes 2013(41)73–79.

12.

Roy

. The outranking approach and the foundations of ELECTRE methods. Thero Decis 1991; 31: 49–73.

13.

Vahdani

Jabbari

AHK

Roshanaei

et al.

Extension of the ELECTRE method for decision-making problems with interval weights and data. Int J Adv Manuf Tech 2010; 50(5–8): 793–800.

14.

Corrente

Greco

Slowinski

. Multiple criteria hierarchy process with ELECTRE and PROMETHEE. Omega 2013; 41(5): 820–846.

15.

Kuo

CFJ

Huang

Lin

et al.

Optimal injection parameters for multiple qualities in poly lactic acid/nano mica composites. Text Res J 2014; 84(14): 1489–1505.

No.	1	2	3	4	5	6	7	8	9	sum
1	–	0	1	0	0	0	0	0	0	1
2	1	–	1	0	0	0	1	1	1	5
3	0	0	–	0	0	0	0	0	0	0
4	1	0	1	–	0	1	1	1	1	6
5	1	1	1	1	–	1	1	1	1	8
6	1	0	1	0	0	–	0	0	0	2
7	1	0	1	0	0	0	–	0	0	2
8	1	0	1	0	0	1	1	–	1	5
9	1	0	1	0	0	1	1	0	–	4

No.	1	2	3	4	5	6	7	8	9	sum
1	–	0	1	0	0	0	0	0	0	1
2	1	–	1	0	0	0	1	1	1	5
3	0	0	–	0	0	0	0	0	0	0
4	1	0	1	–	0	1	1	1	1	6
5	1	1	1	1	–	1	1	1	1	8
6	1	0	1	0	0	–	0	0	0	2
7	1	0	1	0	0	0	–	0	0	2
8	1	0	1	0	0	1	1	–	1	5
9	1	0	1	0	0	1	1	0	–	4

No.	1	2	3	4	5	6	7	8	9	sum
1	–	0	1	0	0	0	0	0	0	1
2	1	–	1	0	0	0	1	1	1	5
3	0	0	–	0	0	0	0	0	0	0
4	1	0	1	–	0	1	1	1	1	6
5	1	1	1	1	–	1	1	1	1	8
6	1	0	1	0	0	–	0	0	0	2
7	1	0	1	0	0	0	–	0	0	2
8	1	0	1	0	0	1	1	–	1	5
9	1	0	1	0	0	1	1	0	–	4