A review of synthetic latices in surgical gloves.
There are currently two types of synthetic polyisoprene. The most widely used industrial process to produce high-cis polyisoprene is Ziegler-Natta (ZN) polymerization; this process can yield synthetic polyisoprene with cis contents above 96%. The alternative is anionic polymerization, which leads to cis contents above 90%. The latter process is uniquely employed by Kraton Polymers in its Belpre, OH, facility, and branded under the name Cariflex polyisoprene products. A unique feature of Cariflex IR is its extreme purity and consistently high quality, which greatly simplifies manufacturing and quality control (ref. 1). An overview of the different types of isoprene rubber is presented in table 1.
For the manufacture of dipped goods, such as surgical gloves, the rubber needs to be available as an aqueous emulsion, while IR production is solvent-based. Hence, for ZN-IR and Cariflex IR, a secondary step is required which involves the emulsification of the rubber in solution into an oil-in-water system. The so-called latex production process is designed to lead to synthetic latices that mirror the key properties of NR latex, as is presented in table 2 for Cariflex IR.
Surgical gloves produced from either natural rubber or synthetic rubbers have different physical standards which they have to meet according to ASTM D3577-01a (ref. 2), summarized in table 3.
The ASTM requirements are mostly designed to guarantee sufficient protection of the surgeons and patients, but not so much to ensure comfort during use. An indication of comfort is given by the modulus at low elongation. Accordingly, in our study we have also investigated the moduli at lower elongation. Replacement of natural rubber latex surgical gloves by synthetic alternatives has caused in the past some concerns regarding comfort and protection. For example, Thomas et al. reported at the 2010 Annual Meeting of the American Academy of Orthopaedic Surgeons that latex-flee gloves may be more likely to perforate than latex gloves, especially in arthroplasty (ref. 4). This was based on a field study involving only two different hospitals, two types of latex gloves and three types of latex-free gloves.
The current study is aimed at systematically and quantitatively evaluating various types of commercial surgical gloves. It involves testing mechanical properties of four types of NR gloves three types of anionic IR gloves, two types of ZN-IR gloves and three types of polychloroprene gloves. Mechanical properties measured include those related to protection (tensile strength, tear strength and puncture resistance) and comfort (small deformation modulus, modulus at 500% elongation and hysteresis).
Materials and methods
Twelve different surgical gloves from different manufacturers were obtained, all size 7 1/2. Four of these were manufactured from natural rubber (NRA, NR-B, NR-C and NR-D), three were prepared from Cariflex anionically polymerized polyisoprene IR401 latex (AnIR-A, AnIR-B and AnIRC), two were made from Ziegler-Natta polyisoprene latex (ZN-IR-A, ZN-IR-B) and three from chloroprene (CRL-A, CRL-B and CRL-C).
Thickness of the gloves was determined according to ASTM-D3577 (ref. 2), at the middle finger, at the palm and at the cuff. Thicknesses were measured using a Marcator 1086 gauge, equipped with a flat probe.
[FIGURE 1 OMITTED]
ASTM D412 (ref. 3) was followed in measuring tensile strength, different moduli and elongation at break. Dumbbells type C were cut from the gloves; one dumbbell from the palm, another one from the back of the hand, both in the direction of the fingers. For all tests, an Instron type 3365 tensile bench was used, either with or without a long-range traveling extensometer.
Because for this type of elastomeric article it is impossible to determine the real Young's modulus reproducibly, we measured the modulus in the range of 5% to 15 % extension by linear regression over this region and called this value small deformation modulus. In this way, it was possible to obtain a reproducible measure for the modulus at small strains, which seems to be a valuable measure for the use of surgical gloves (important e.g., for finger movements).
[FIGURE 3 OMITTED]
The different polymers used to prepare the surgical gloves will have different strain induced crystallization behavior. When a material crystallizes and melts during a cyclic deformation, the material produces a substantial hysteresis loop: The hardening induced by crystallization during stretching is re covered during unloading. The area inside the loop represents the crystallization energy. Hysteresis curves were recorded by measuring the stress-strain curve up to a certain deformation at 500 mm/min., and subsequently recording the same curve when returning to zero strain, again at 500 mm/min. This procedure was repeated on the same sample for increasing maximum strains in sequential loops.
[FIGURE 2 OMITTED]
Tear strength of the samples was measured according to ASTM-D624 (ref. 5). Because it is known that use of the trouser shaped die often results in so-called knotty tearing, only the v-shaped specimen was analyzed. Samples were cut from a glove in the same way as was done for the tensile measurements: one from the palm of the hand, the other from its back. Tear strengths were measured at a grip separation speed of 500 mm/min.
Puncture resistance was determined according to ASTM F1342 (ref. 6), and according to ASNZL 4179 (ref. 7). In the first test, a puncture probe as shown on the left side of figure 1 was used; in the second, the probe shown on the right side of figure 1. The samples were kept between two metal plates, with chamfered holes. In the case of ASTM F1342, the diameter of the holes was 6.4 mm; for ASNZL 4179, it was 10 mm.
Except for three types, thickness &all gloves was on the order of 200 micrometers. AnIR-B and ZN-IR-B were on the order of 250 micrometers, CRL-A was about 175 micrometers. All gloves were thickest at the finger, and thinnest at the cuff.
An impression of the crosslink density was obtained by swelling a small circular disk in toluene. The average molecular mass between two crosslinks was found to be between 6,500 and 8,500 for all samples. The degree of swelling decreases as the crosslink density increases, and at values as measured for the gloves, it is experimentally almost impossible to determine statistically significant differences (ref. 8). The conclusion that can be drawn from the results is that all gloves are well crosslinked.
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
Tensile strength, moduli and elongation at break
Tensile measurements were made following the ASTM D412 standard cited in the surgical glove standard. Figure 2 contains typical example tensile curves of four of the gloves tested. As can be seen, the tensile forces are all of the same order of magnitude for all samples (later in this report the NR sample will be discussed). The anionically polymerized polyisoprene has the highest elongation at break, the natural robber sample the lowest. The curves indicate that the modulus at small strain is lowest for the anionically polymerized polyisoprene, followed by Ziegler-Natta polyisoprene and natural rubber. Chloroprene has a higher modulus. We will look at this aspect in more detail later in this section.
The tensile strengths of the different gloves are shown in figure 3. The left columns give the results as measured using the long traveling extensometer to record the stress-strain curve. It is remarkable that the tensile strengths of the NR gloves all were quite low, below the specification for unaged samples. The standard deviation of the NR results was also significantly larger as compared to the other types. The AnIR, ZN-IR and CRL glove types were all well within the specification for unaged synthetic surgical gloves.
It turned out that the reason for the unexpectedly low figures for the NR gloves was in the use of the extensometer. The right columns in figure 3 represent the tensile strengths as measured without the extensometer. As can be seen from the comparison of the results obtained with and without the extensometer, only the NR samples are influenced by its use. This may indicate that NR gloves are susceptible to small disturbances while under stress.
For the AnIR gloves, sample AnIR-B is somewhat stronger than the other two, which correlates with their thickness. For the chloroprene gloves, CRL-A is stronger than the others, but this glove is the thinnest of the three.
When tested without the extensometer, all of the surgical gloves tested met the minimum tensile strength criteria for unaged samples set forth in the ASTM standard: 24 MPa for NR, 17 MPa for synthetic gloves.
It was remarkable to observe that the break pattern of the NR and ZN-IR gloves was different than the pattern of the AnIR or the CRL types. This is shown in figure 4. It is clear that the NR and ZN-IR types show an irregular break pattern, whereas the other types give a rather straight cut, perpendicular to the direction of strain. The difference in break pattern for NR and ZNIR can be an indication that their tearing behavior is different than AnIR and CRL. The tearing propagation of the latter two may be faster.
The other properties of the unaged gloves tested, complying to ASTM D3577, are the modulus at 500% and the elongation at break. The elongation at break is also influenced by the use of the extensometer (figure 5); with the extensometer the elongation at break is smaller. This can be explained by the fact that the dumbbell broadens into a shoulder in the neighborhood of the grip, so here it can more than proportionally elongate.
When comparing stress-strain curves measured with and without an extensometer, the latter seems to be extended towards higher strains, as compared with the former.
The modulus at 500% is shown in figure 6, both using the extensometer and without. When looking at the results obtained using the extensometer, all the NR gloves fail the criterion of modulus 500% < 5.5 MPa. Of the synthetic gloves, ZN-IR-B and CRL-B fail the specification for synthetic gloves <7 MPa. The results obtained without using the extensometer indicate that both NR and synthetic gloves are within their specifications.
The difference in modulus at 500% as obtained with or without an extensometer can be explained by the effect described above. The stress-strain curve appears to be extended towards larger strain values when using the extensometer. Therefore, the stress at 500% deformation as indicated by the measurement without an extensometer is actually the stress at a lower deformation when using the extensometer.
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
The fact that the modulus at 500% for NR is so much higher when measured with an extensometer than without it can be explained by looking at figure 2 again. From this graph it is clear that NR and CRL show lower values for the elongation at break than AnIR. This implies that the stress-strain curve shows the steep increase already at lower elongations at break. In this steep region, a small difference in elongation can make a large difference in force (and therefore modulus).
The question is which are the best/more accurate data, with or without an extensometer? In principle, the ones measured with the extensometer would be more reliable, because with this device only the isotropically deforming part of the dumbbell is taken into account. However, because the extensometer is attached to the dumbbell mechanically, by a tiny force, it may disturb the measurement of the tensile strength. Just as moduli are more accurately determined with the use of the extensometer, tensile strength may be measured more reliably without it. Elongation at break is, in principle, more accurate using the extensometer, but for natural rubber its use may induce premature break.
It is important to note that, except when he is slipping his hands into the glove, a surgeon will probably never reach, during common
use, the 17 or 24 MPa tensile strengths required for synthetic and natural rubber based gloves, respectively. The maximum stress at 500% elongation may also never be realized, as typical use (bending fingers) only elongates the glove by 50 to maybe 100% (use region will be defined as the properties up to approximately 100%). Thus, while the physical properties dictated by the ASTM standard are required for a glove to be labeled and used as a surgical glove, these are not indicative of glove comfort.
In general, the comfort of a glove seems to be associated with its stiffness. Since the feel of a glove involves small deformations, the most likely parameter to characterize softness is the stiffness at small elongations. Ideally, the Young's modulus would be the measure of choice to compare materials. Figure 7 gives the stress-strain curve at low deformations for the four types of material. From this graph, it becomes clear that it is almost impossible to obtain a reliable value for the Young's modulus. Therefore, we decided to compare moduli at small deformations (between 5 and 15 %) and call this the small deformation modulus.
[FIGURE 8 OMITTED]
[FIGURE 9 OMITTED]
The moduli at very small deformation, at 25% and at 100% deformation, are shown in figure 8. The results in figure 8 show that anionic polyisoprene is softer than natural rubber and Ziegler-Natta polyisoprene. Conversely, chloroprene is the stiffest (highest Young's Modulus) and is generally considered the least comfortable.
The degree of strain-induced crystallization of the glove in tension was measured via a hysteresis method. When a material crystallizes and melts during a cyclic deformation, the material produces a substantial hysteresis loop (see figure 9 for raw plots). During the first cycle, there is typically both non-recoverable and recoverable hysteresis; however, during the second and subsequent cycles, most hysteresis is due to crystallization during stretching followed by melting during retraction. In other words, the hardening induced by crystallization is recovered during unloading. The area inside the loop represents the crystallization energy. Thus, hysteresis experiments can be used as a measure of crystallinity during stretching, and to identify the strain at which crystallization begins.
Figure 9 shows typical examples of the hysteresis curves for the four glove types. It is clear that NR and CRL materials show the largest crystallization energy, with ZN-IR at an almost equal value. The gloves prepared from AnIR show the least hysteresis. This is in accordance with the cis contents of the materials: Anionically polymerized polyisoprene typically shows cis percentages in the low 90s, whereas NR and ZN-IR have percentages in the upper nineties. The microstructure of the CRL polymer will mainly be trans.
Figure 10 shows the hysteresis loss values for the anionically polymerized polyisoprene gloves, as well as gloves manufactured from NR, Ziegler Natta polyisoprene and chloroprene rubber at different elongations. The data shown are based on the plots of the hysteresis data. This experiment differentiates the materials clearly by their base polymer type. The CRL and NR gloves show the most hysteresis, the ZN-IR gloves less, and then the AnIR. The curves for the AnIR gloves show only a small crystallization induced hysteresis loop. Furthermore, this loop only becomes visible at higher elongations as compared to other materials.
At elongations of 50-100%, AnIR is the only material not showing strain-induced crystallization. Thus, the graphs suggest that only for AnIR the danger is absent that local crystallization in a glove during use negatively affects comfort.
Tear strength of the gloves was measured according to ASTM D624 using the V-shaped die. Results are shown in figure 11. For the NR samples, the results seem to be very similar: All four have values around 50 kN/m, albeit that NR-C is somewhat lower. All three AnIR samples consistently show values of about 25 kN/m. For the ZN-IR samples the variation is larger: ZN-IR-B shows a tear strength of about 50 kN/m, very similar to NR, whereas the value obtained for ZN-IR-A is considerably lower. Because the microstructure of ZN-IR is very similar to that of NR, beforehand it was expected that mechanical properties, including tear strength, would be very similar as well.
[FIGURE 10 OMITTED]
[FIGURE 11 OMITTED]
The CRL samples also show a somewhat mixed picture, with CRL-A on the same level as the AnIR samples, and CRL-B and CRL-C having lower tear strength values of about 20 kN/m.
Bueche (ref. 9) states that tear strength is not dependent on tensile strength as such, but it is the energy furnished to the rubber in extending it to its maximum elongation that is important. So, not only the strength of the material, but also its modulus and elongation are involved. However, calculating energies from the stress-strain curves revealed that they are very similar for all glove types. The AnIR types even gave slightly higher values than the others.
The puncture resistance of the gloves was tested in two ways, using two differently shaped tips, and two holes having different diameters.
Results from the ASTM F1342 tests are shown in figure 12. All puncture forces measured are between 3.7 and 5.5 N and vary within all product groups. According to the ASTM F1342, when the elongation exceeds 20 mm, the test should be stopped to prevent damage to the test setup. However, because we took precautions for this, we recorded the puncture force at puncture.
The elongation at which puncture occurs is very similar for almost all gloves, except for two AnIR gloves, which show a larger elongation at puncture.
[FIGURE 12 OMITTED]
[FIGURE 13 OMITTED]
The best way to look at puncture resistance may be to compare the energy it takes to puncture the material. The values for the two types of puncture tests carried out are given in figure 13. From these results it is clear that two out of three of the AnIR and one of the CRL gloves are able to absorb the most energy before they become punctured. The results of figure 13 indicate that it is possible to manufacture gloves from Cariflex IR401 having a better possibility to absorb puncture energy.
Summary and conclusion
Twelve different gloves, produced from four different types of material were evaluated with respect to their mechanical behavior.
It was shown that all gloves met the ASTM standard for surgical gloves. However, when the extensometer was used, tensile strength of the natural rubber gloves was below the specification for all four samples. Thus, NR seems to be susceptible to small disturbances when under stress. AnIR and CRL suffer from easier tearing propagation than NR and ZN-IR, but sustain higher or equivalent puncture energy, which may be related to tear initiation. Based on these data, we infer that all glove types studied offer comparable mechanical protection.
All four types of material show hysteresis when cycling through force-extension loops. For the anionically polymerized polyisoprene, the hysteresis is the least, indicating strain induced crystallization is less. The hysteresis graphs suggest that only for AnIR the danger is absent that local crystallization in a glove during use negatively affects comfort. Small strain modulus is lowest for AnIR, medium for NR and ZN-IR, and highest for CRL. In summary, we infer that gloves prepared from AnIR offer better comfort.
(1.) P. Henderson, From Isoprene Monomer to Synthetic Polyisoprene Latex and its Uses (2001).
(2.) ASTM D3577-01a, Standard specification for rubber surgical gloves.
(3.) ASTM D412, Standard test method for vulcanized rubber and thermoplastic elastomers--tension.
(4.) Thomas S., Aldlyami E., Gupta S., et al. Clinical performance of latex-free gloves: Is it safe for arthroplasty surgery? Paper #290. Presented at the 2010 Annual Meeting of the American Academy of Orthopaedic Surgeons. March 9-13, 2010. New Orleans.
(5.) ASTM D624, Standard test method for tear strength of conventional vulcanized rubber and thermoplastic elastomers.
(6.) ASTM F1342, Standard test method for protective clothing material resistance to puncture.
(7.) ASNZL 4179, Single-use sterile surgical rubber gloves-specification. Appendix YY, Requirement and test method for glove cuff rupture resistance.
(8.) J.L. Valentin, J. Carretero-Gonzalez, L. Mora-Barrantes, W. Chasse and K. Saalwachter, Maeromolecules, 41, 4,714 (2008).
(9.) F. Bueche, "Physical properties of polymers." Interscience Publishers (1962).
Table 1--polyisoprene production--different processes give different products Natural rubber Ziegler-Natta IR Cariflex IR (anionic 1R) No catalyst Titanium/aluminum Alkyl-lithium catalyst catalyst Wide molecular Wide molecular weight Narrow molecular weight distribution weight distribution distribution 98+ % cis content 96+ % cis content 90+ % cis content High gel content High gel content Intrinsically no gel Contains natural Catalyst residuals Low impurity level impurities Produced as Produced in organic Produced in organic emulsion solution solution Table 2--typical properties of Cariflex IR and NR Latex Cariflex IR NR Total solids content (wt. %) 65 Similar Total rubber content (wt. %) 64 Similar pH 11 Similar Ammonia (wt. %) 0 0.2-0.8 Average particle size (micron) 1.1-1.5 0.9 Mechanical stability (sec.) >1,500 >900 Brookfield viscosity (mPa.s) <150 Similar Table 3--physical standards for surgical gloves according to ASTM D3577-01a--the physical properties are measured using ASTM D412 (ref. 3) Type Before aging Minimum Minimum tensile ultimate strength elongation MPa % Natural rubber 24 750 Synthetic rubber 17 650 Type After accelerated aging Maximum Minimum Minimum stress at 500% tensile ultimate elongation strength elongation MPa MPa % Natural rubber 5.5 18 560 Synthetic rubber 7.0 12 490
|Printer friendly Cite/link Email Feedback|
|Author:||Krutzer, Bert; Ros, Marianne; Smit, Joris; de Jong, Wouter|
|Date:||Nov 1, 2012|
|Previous Article:||The rubber economist.|
|Next Article:||Natural rubber-silica combinations for low rolling resistance truck tire treads.|