Preliminary Study on Effect of Chemical Composition Alteration on Elastic Recovery and Stress Recovery of Nitrile Gloves

Nitrile gloves are widely used in the medical and automobile field due to its superiority in hypo-allergic component and chemical resistance over natural latex gloves. However, poor elastic recovery of nitrile glove to compressive force also creates an aesthetic issue for customers with high levels of wrinkling after removing from glove box. This paper demonstrates the preliminary study on the varies chemical composition such as crosslinking agents, sulphur and zinc oxide, the accelerator agent added during curing process, and the rubber filler Titanium Dioxide, on the elastic recovery and stress relaxation in nitrile gloves manufacturing. These chemical were studied at different concentration level comparing the high and low level versus the normal production range. Due to the inconsistency in the analysis technique on the surface imaging, the elastic recovery result was unable to be quantified and was not conclusive at this point. The cross linking agents, sulphur and zinc oxide, and the accelerator agent, played a significant role in the mechanical strength of the gloves. Increment of these chemicals result in higher tensile strength, but a reduction in the elasticity of the materials in which causes a lesser elongation at break percentage for the gloves. Both cross-linkers demonstrate different behaviour where higher sulphur content, provide higher stress relaxation (SR%) yet zinc oxide shows otherwise.


Introduction
Nitrile gloves are synthetic latex gloves also known as Nitrile Butadiene Rubber (NBR) gloves. It is a copolymer derived from 1,3 Butadiene (Petroleum derived) and Acrylonitrile instead of renewable latex source such as Natural Rubber (NR) gloves [1]. NBR gloves are preferred over NR gloves as medical examination due to its superiority in oil resistance, chemical resistance and mechanical strength over NR gloves [2]. Up to 17% of workers in the medical industry suffers from latex allergy which causes itching and drying of skin with similar traits to rashes. This makes nitrile examination glove to mainstay option for medical use.
Several terms are commonly used to describe mechanical properties of rubber, these qualities determine the type and grade of rubber found in the market. Tensile strength is defined as the amount of force (N) per square meter (m) needed to elongate a rubber sample to the point of failure (rupture). Stress Relaxation is defined as the decay of stress under constant strain for a certain period of time due to internal molecular rearrangement (Deformation) to achieve equilibrium with applied force. Rubber with higher tensile strength would require higher load (N) of stress to cause material failure. Elongation (%) is the percentage of stretch of rubber material from its initial state presented in percentage, elongation at break (%) is the measurement of stretch at breaking point of rubber sample. Rubber with high elasticity would display higher elongation (%) when stretched compared to stiffer samples. Elastic modulus also known as Young's Modulus is a measurement of stiffness of a certain material, which is defined as the amount of force exerted at a certain elongation. As such, stiffer or harder rubber samples would display a higher elastic modulus at a certain elongation compared to softer and elastic samples [3]. Rubber Creep is defined as the permanent set of plastic strain experienced by a material under constant strain. Basically rubber samples that have high stress relaxation properties have higher creep when subjected to a certain stress [4] Issue with Nitrile Glove products Physical properties between NBR and NR latex have contrasting difference due to their difference in molecular constituents. Natural latex (NR) rubber generally have superior elasticity, resilience and softness compared to NBR, which in nature has higher stiffness, poor elasticity, poor resilience but better tensile strength. The unique characteristics of NBR is linked to the presence of Acrylonitrile (ACN) monomer in its structure, increase in ACN content increases polarity of NBR molecule and glass transition temperature (Tg) properties. Although NBR gloves posses better mechanical properties compared to NR gloves, the stiffer and less elastic nature of NBR gloves do bring about several issues. One issue of NBR gloves is the poor elastic recovery (issue 1) from compressive stress, this results in NBR gloves easily forming surface wrinkles when subjected to compressive force when folded. The other issue regarding NBR glove is its stress retention when worn by any user, the low stress relaxation (Issue 2) property of NBR gloves makes it less comfortable when donned.
This project is a research collaboration between Top Glove Sdn Bhd and Taylor's University to address the two stated issues regarding their Nitrile glove products. The objective is to look into effect of chemical composition optimization to improve stress relaxation and elastic recovery properties. Customer feedback for Top Glove's nitrile glove products has seen customers commenting on the high degree of wrinkling when removed from glove box, which is aesthetically unpleasing as well as causing inconvenience for customers to "open" the gloves up from wrinkled state. The poor stress relaxation of Nitrile gloves makes it less comfortable when donned for long periods as nitrile glove would not conform to the user's hand. Therefore, a glove that has higher stress relaxation properties would experience deformation (loss of stress) over time and conform to shape of user's hand giving a better comfort when worn for long durations.

Latex compounding chemicals selected for study
This study aims to look into the effect of chemical composition optimization of four chemicals to enhance nitrile glove elastic recovery from compression and stress relaxation. The four chemicals stated are as stated below: 1. Sulphur (Vulcanizing agent) 2. Zinc Oxide (Vulcanizing activator) 3. Accelerator (Vulcanizing catalyst) 4. Titanium Dioxide (Filler) Sulphur (S) is the most common vulcanizing agent used in the rubber industry, where it provides sulphur cross-links between long chain polymer molecules which would determine the produced glove properties. Zinc Oxide (ZnO) act as a vulcanizing activator where it would enhance cross-linking between polymer chains and form ionic cross linkage between polymer molecules. Addition of ZnO enhances vulcanizing efficiency and acts as an activator for sulphur crosslinking [5]. Accelerators are also added into NBR latex as vulcanizing catalyst to greatly increase speed of vulcanization with process taking a much shorter time while reducing the use of sulphur in the system [6]. Composition of these three chemicals involved in rubber curing would hold an crosslink density of produced rubber and subsequently the mechanical properties of produced glove [7]. Titanium Dioxide (TiO2) acts as a rubber filler which reinforces rubber structure for better mechanical properties. Preliminary study will be conducted to test significance of composition change of each chemical towards stress relaxation and elastic recovery of produced latex film. Three of the four chemicals with most significant effect will be studied for optimization to obtain resultant latex with best stress relaxation and elastic recovery properties.

Research Objectives
1. (Preliminary Study) To select three out of the four target chemicals with most significant effect on stress relaxation, tensile strength and elastic recovery properties. 2. To optimize chemical composition of target chemicals to achieve nitrile latex with best stress relaxation and elastic recovery property. 3. To achieve enhanced properties while keeping mechanical properties of nitrile latex within industry standards ASTM D6391 (

Methodology
Chemical composition for significance test Based on the four target chemicals discussed, composition of chemicals is adjusted to extreme highs and lows to test significance of each chemical towards targeted response (stress relaxation and elastic recovery). Table 2.1 shows the composition of chemicals categorized into three levels. Medium level composition represents normal phr values used in nitrile glove production. "Phr" is an industry standard of denoting chemical ratio based off 100 parts of rubber. when one chemical has its composition varied between high and low values (denoted using bold font), all other chemicals phr value is kept at the middle (normal) range to obtain an accurate comparison between produced latex. With this variation, a total of nine batch of latex samples is compounded for single factor study as shown in Table 2.2 below.  Table 4 is not included for confidentiality reasons* Chemical weight (g) = chemical phr x Latex TSC ሺ%ሻ x Latex weight (g)

*Please note that TSC and Phr information for chemicals in
Chemical TSC (1)

Latex blend compounding
Latex compounding involves constant stirring of latex and added chemicals for a period of 24-48 hours to produce a homogenized solution. This step is important in order to avoid settlement of high density solid particles, which may cause uneven properties throughout surface of produced latex. For this study, several conditions are set to be constant for all compounding batches to standardize the compounding process. 24 hour period for latex compounding is chosen based on studies done by Ruslimie et.al [8] showing no increment in crosslink density (mole/g) and tensile strength (MPa) for produced latex with pre-vulcanization time over 24 hours.

Elastic recovery from compressive stress
An important note is that there is no industrial standardized method of measurement of surface wrinkles through surface imaging or elastic recovery from compressive force for thin rubber films such as glove samples. Method of surface imaging and surface wrinkle analyzation discussed below represents a new and untested method developed through collaborative efforts between Ms. Eunice Phang Siew Wei (Program director/Senior lecturer of Taylor's University) and industrial supervisors at Top Glove Sdn Bhd. This newly developed method can be categorized into 3 stages: Compressive folding of latex sample (Uniform and non-uniform folding) II.
Surface image scanning using a scanner III.
Analyzation and calculation of surface wrinkles (%) through MATLAB Produced latex samples are folded using two different methods (Figure 2.4.1) to compare which method provided substantial improvement in elastic recovery after a certain period. Surface imaging for samples are scanned for surface morphology analyzation using MATLAB to determine which method is best suited.  Tensile strength test Dumbbell shaped test samples were gripped on onto clips of the Universal tensile machine (Figure 2.2 Test results as shown in Table 3.1 were clearly non-repeatable and non-sensible. Although MATLAB was not able to differentiate elastic recovery in 10 minutes, the minor differences can be seen through visual judgement based on scanned images shown in Error! Reference source not found. below. There is a certain amount of recovery in wrinkling depth between both samples. The scanned image after 10-minute of recovery (right) is bigger in terms of surface area compared to sample image at 0minute due to the minor elastic recovery in wrinkle depth causing latex sample to slightly "expand" in size as it recovers. Wrinkled samples would display higher degree of elastic recovery after longer amounts of time, but that would beat the purpose of the objective, as end users were dissatisfied with the high number of wrinkles when gloves were immediately removed from glove box (Elastic recovery in short period) Hence the issue with results being non-sensible (negative values) and non-reproducible can be summarized in 2 major factors: 1. MATLAB analyzation limitations As discussed with the industrial supervisor who wrote the coding, MATLAB was only able to contrast wrinkled (white) and unwrinkled regions (black) as shown in Figure  2.2 and calculate total surface area of wrinkled regions (white) over total selected surface area. But based on visual differences shown in Figure 3.2, regions of wrinkle lines and regions remains largely the same for samples at 0-minute and 10-minute. Only a change in depth of wrinkles were observed. Hence MATLAB was not able to distinguish the differences between both images making results insignificant. Future improvements of coding may allow or wrinkle depth to be differentiated using MATLAB.
2. Inconsistency in manual adjustment for selected surface area As shown in Figure 2.2, selection of surface area is done by manually adjustments of the blue indication box. This adds in factor or inconsistency when surface area of 0minute sample and 10-minute sample are selected. Since wrinkle regions show no difference in MATLAB between both images, difference in selected area may cause an increment or decrease in resultant elastic recovery (%). This explains the negative values of elastic recovery (%) shown in Error! Reference source not found.. Future improvements can be done in coding to automatically identify and select area of latex sample to avoid such consistency.

Significance test for stress relaxation and mechanical strenght
Target chemicals (Sulphur, Zinc Oxide, Accelerator, Titanium dioxide) have phr adjusted beyond normal production range (Refer to Table 2.1) to select three chemicals with most significant change in stress relaxation (%) response. Table 3.2 below shows stress relaxation (%) results of each chemical variation. All data shown for stress relaxation and tensile strength were averaged out between data obtained from 4 test samples of the same composition batch. Table 3.3 below shows the tensile strength, elongation at break (%) and elastic modulus at 300% elongation data for the same samples. Titanium dioxide (TiO2) which acts as a reinforcement filler, did not have significant effect in stress relaxation (%) when concentration is adjusted to extreme highs and lows. The other three target chemicals (Sulphur, Zinc Oxide, Accelerator) were directly involved in rubber curing process, change in concentration in these chemicals would have a significant impact in crosslink density of produced latex. Stress relaxation (%) between High concentration (3.0 phr) and low concentration (0.2 phr) only resulted in a difference of 0.47 MPa. The insignificant difference for high and low samples of TiO2 in stress relaxation (%) can be backed up by mechanical test data shown in Table 3.3, which shows the least difference across the board for tensile strength, elongation at break (%) and elastic modulus at 300% elongation. Data shown in this table further proves that a direct influence in crosslink density can effect mechanical strength and overall stiffness of a latex sample.
An additional observation can be made in Table 3.2, stress relaxation (%) high and low concentration of Zinc Oxide (ZnO) shows an exceptionally higher difference compared to other chemicals, with difference of 13.30% in stress relaxation. This clearly shows that ZnO is the most significant factor involved in enhancing stress relaxation in produced latex. This may hint that ionic crosslinking in rubber molecules can significantly impact mechanical properties of produced rubber. By referring to results in Table 3.3, elongation at break (%) High and low Zinc Oxide samples show the highest difference of 213.68% between samples from both batch. The same trend is seen with elastic modulus at 300% elongation. High ZnO samples had exceptionally high elastic modulus at 22.17 MPa while low ZnO samples were exceptionally low at 2.09 MPa. This meant that low ZnO samples were exceptionally soft, with much lesser load (N) needed to stretch samples to 300% elongation. Based on discussions with industrial supervisors on the relation between all these data, a summarization between the relationship of mechanical properties can be summarized below in Table 3.4 below: Results from preliminary study of significance in effect of chemicals can conclude two major statements: 1. Titanium dioxide (TiO2) was the least significant chemical towards stress relaxation (%). It will be excluded for full factorial design and composition optimization.
2. Zinc Oxide (ZnO) is the most significant factor affecting stress relaxation (%) and overall softness of produced latex.

Conclusion
As it stands, elastic recovery measurement through surface imaging using scanner shows that the method is not suited for this project as results obtained are inconsistent and non-repeatable. MATLAB analyzation of surface wrinkle was only able to determine the number of wrinkles / wrinkled regions of selected area but unable to distinguish the depth of surface wrinkles on nitrile latex surface. Elastic recovery of wrinkled films had recovery of wrinkled depth but little to no recovery from number of wrinkles after 10 minutes of recovery period. Hence, MATLAB was unable to distinguish difference between sample image at 0 minute and 10 minute. Manual selection of analysed surface area through manual adjustment also caused inconsistency in selected areas for 0 minute and 10 minutes samples, resulting in some samples yielding negative elastic recovery values. Titanium Dioxide (TiO2) showed least significant changes for SR% and mechanical properties between high-low concentrations and between normal formulation samples. Increment in sulphur and accelerator concentration caused an increment in SR%, possibly due to increased sulphur crosslinking in samples. But increment of Zinc Oxide caused decrease in SR% due to ionic crosslinking between rubber molecules. Tensile strength results show an expected outcome with an increment in mechanical strength with higher concentrations of sulphur, accelerator and zinc oxide due to higher density crosslinking. But enhanced mechanical strength had a resultant effect of decreased elasticity of produced samples, with elongation at break (%) showing decreased value with higher concentrations. Sulphur, accelerator and Zinc Oxide are chosen for full factorial study for chemical composition optimization for next phase of the study.