Abstract
Pre-filled syringes/cartridges as primary packaging for parenterally delivered biopharmaceutical liquids consist of multiple components, including containers made of glass or plastic, and stoppers/plungers and disk seals (septa) made of rubber materials. Cracking of rubber components may be cosmetically unacceptable and in extreme cases may compromise enclosure integrity. The purpose of this study was to investigate the root cause of septum cracking and evaluate parameters/solutions to delay or prevent cracking from occurring. Custom-made chambers capable of tightly controlling ozone levels were assembled to deliberately create septum cracks. Cracks were qualitatively assessed by optical microscopy and quantified using image analysis by ImageJ. The results confirmed that ozone attack is the root cause of septum cracking during storage, and the stress—the result of crimping on the glass cartridge by the aluminum lined seal—made the septum particularly vulnerable to ozone attack. Ozone concentration as low as 10–40 ppb (levels routinely detected on a busy street) could crack the stressed septum in hours while days of ozone exposure at 50 ppm could not cause the unstressed septum to crack. Under ozone attack cracks initially grow in length and width uniformly across the stressed area and then stop progressing, perhaps due to residual stress release. Although the use of impermeable barriers could prevent cracking completely, this study suggested that any form of packaging barriers, including a highly permeable Tyvek® sheet, could postpone cracking by slowing down ozone diffusion and convection. We demonstrate that simple double packaging—placing the Tyvek®-lidded blister tray in a cardboard carton—could sufficiently protect the stressed septum for years in a surrounding environment with ozone at normal indoor levels (≤2 ppb).
LAY ABSTRACT: Pre-filled syringes/cartridges as primary packaging for parenterally delivered biopharmaceutical liquids contain multiple components, including a disk seal (septum) made of rubber materials. Cracking of rubber components may be cosmetically unacceptable and in extreme cases may compromise enclosure integrity. The septum, if not appropriately packaged, might crack under uncontrolled storage environment. The purpose of this study was to investigate the root cause of septum cracking and evaluate parameters/solutions to delay or prevent cracking from occurring. Custom-made chambers capable of tightly controlling ozone levels were assembled to deliberately create septum cracks. The results confirmed that ozone attack is the root cause of septum cracking during storage, and the stress—the result of crimping on the glass cartridge by the aluminum lined seal—made the septum particularly vulnerable to ozone attack. Ozone concentration as low as 10–40 ppb (levels routinely detected on a busy street) could crack the stressed septum in hours. Although the use of impermeable barriers could prevent cracking completely, this study suggested that any form of packaging barriers, including a highly permeable Tyvek® sheet, could postpone cracking by slowing down ozone diffusion and convection. This investigation will raise awareness of manufacturers of pre-filled cartridge/syringe parenteral products to storage and packaging requirements for the long-term physical stability of cartridge components as small as the rubber septum.
Introduction
Large molecules like monoclonal antibodies, proteins, and peptides need to be delivered via the parenteral route. Pre-filled cartridges and syringes are now the primary container of choice for most parenteral drug delivery systems mainly for reasons that they are safe and user friendly (1, 2). Cartridges are normally inserted into delivery pens and primarily used for multiple-dose proteins such as insulin and growth hormone, including Humulin HumaPen®, NovoLog® FlexPen®, Lantus SoloSTAR® Pen, Nutropin AQ® NuSpin™, and others. For a standard cartridge (Figure 1a), a liquid formulation is filled into a glass tube closed on either side by a rubber plunger and a rubber disk seal (or septum). The septum, tightly crimped onto the cartridge via an open-top, aluminium-lined seal, bulges out at the center to dome or form a mushroom, which is termed septum mushroom hereafter. The higher the mushroom, the tighter the seal and the greater the residual seal force on the septum. It is common to store a pre-filled cartridge in a secondary packaging, for example, a plastic blister tray heat-sealed with a barrier lid made of different materials such as Tyvek®, plastic, aluminum foil, among others (Figure 1b). Finally, the blister pack is placed inside a cardboard carton (Figure 1c) to complete the final package.
It was recently found that pre-filled cartridges placed on Rondo trays (unpackaged or naked) and stored in a cold room developed different levels of cracking across the top surface of the septum mushroom and along the mushroom edges near the aluminum-lined seal. Root cause analysis suggested that the uncontrolled storage conditions such as ozone, light, moisture, and temperature, along with inappropriate packaging, might have contributed to septum mushroom cracking (unpublished analysis). It has been reported by the rubber industry, particularly with tires, that multiple factors can deteriorate rubber, mostly by oxidation via chemical and/or mechanical mechanisms (3). Ozone (O3), however, is the most common damaging factor in rubber cracking; it attacks the double bonds in the diene elastomer chain and results in polymer chain scission (4).
A study conducted by West Pharmaceuticals used an exaggerated ozone concentration of ∼5000 ppm (parts per million) on various rubber formulations and found cracking after as little as 3 min of ozone exposure (5). Other studies also demonstrated that ozone attack can be particularly harsh on rubber specimens under stressed conditions (6–8). As a result, a much lower ozone concentration is still sufficient to attack the surface of the rubber and form cracks in areas of high stress.
Temperature and humidity of the cold room referenced above were routinely monitored and showed no unusual excursions. However, 0 to 70 ppb (parts per billion) of ozone concentration was detected, which is unusually high compared to the typical indoor ozone concentration of ≤2 ppb. It is known that the electrical equipment such as fans, electrical motors, fluorescent lighting (UV), and so on, also present in this cold room could be the source of ozone (3). Because the ozone concentration profile of this cold room matched the pattern of the typical outdoor ozone level (maximum in the late morning and close to zero in the night) (8), these unusual ozone level fluctuations might be related to the construction work outside the cold room. Thus, it is important to pay attention to environmental ozone levels during transportation and storage of pre-filled syringe/cartridges involving stressed rubber components, particularly those packaged in highly permeable barriers.
To evaluate the effects of ozone exposure on septum cracking, we designed an ozone chamber to re-create a storage environment that simulated the conditions of the cold room (Figure 2). This ozone chamber allowed us to evaluate other key parameters such as residual seal force on the septum and air movement. More importantly, we assessed secondary packaging effects on septum cracking, including the use of permeable and impermeable barriers. For comparison, we measured the concentration profile of ozone diffusing through the barrier to compare with theoretical calculations. Cracking is difficult to quantify and is normally reported by how fast the rubber specimen develops cracks as observed with optical microscopy or by the length/width of the biggest crack. In this study we developed a quantitative method for calculating the total crack area via ImageJ analysis of digital photographs of the septum.
Materials and Methods
The cartridge (Figure 1a) used in the study consisted of a 3 mL glass tube (9.3 mm inner diameter × 46.0 mm overall height, Gerresheimer Pharma, Bünde, Germany) sealed with a rubber plunger (Formulation 4432, West Pharmaceuticals, Lionville, PA) on one end and with a rubber septum (7.5 mm outer diameter × 1.95 mm height) on the other end. The septum, made of an isoprene formulation (Formulation 7773/4780, West Pharmaceuticals), was crimped with an open-top, aluminium-lined seal (7.5 mm inner diameter × 5.3 mm overall height). The cartridge was then heat-sealed in a blister pack (Figure 1b) that included two parts: a blister tray molded from polyethylene terephthalate glycol (PETG) plastic and a heat-sealable lid. The lid material was either a permeable Tyvek® (1073B, Du Pont, Wilmington, DE) layered with heat sealant (product code 12147, Amcor, Shelbyville, KY) or an impermeable aluminium foil sheet (product code 15127, Amcor). To complete the package, the blister pack was placed in a cardboard carton (Figure 1c).
Ozone Chamber and Control System
We fabricated and assembled an ozone chamber (Figure 2) capable of reproducibly controlling a wide range of ozone concentrations, from 10 ppb to 100 ppm, for ozone exposure investigation. The heart of the chamber is an ozone generation and monitoring system. Ozone was produced by an ozone generator (AC-500G, Ozone Solutions, Hull, IA) via in-house pressurized air and fed into the chamber. The ozone monitor (AC-106L, Ozone Solutions), operating in the range of 1.5 ppb to 100 ppm, detected and recorded ozone concentration based on a pre-set data logging frequency. The monitor then sent the signal via a relay (R-25, Ozone Solutions) to the generator, which could be switched on or off if ozone concentration was below or above the target ozone concentration, respectively.
Lined Seal Crimping Process
After the cartridge was filled with product, a lined seal assembly (a septum and an open-top aluminum cap) was placed on the top of the glass cartridge. Crimping was performed using a custom-made crimping fixture where the cartridge sat on a stand at a slight angle with the lined seal leaning against a stationary wheel. The cartridge was rotated along the wheel to create a closure on the lined seal. The tightness of the seal was controlled by two factors, the position of the knife height (a length that measures from the top of the crimping fixture to the bottom of the cartridge neck) and the compression pressure (crimping pressure). In this study, unless specified otherwise, all lined seals were crimped using a compression pressure of approximately 4 bar (i.e., 58 lbs of force) applied from the bottom of the cartridge and a knife height of approximately 5 mm.
Mushroom Height Measurement
The distance between mushroom climax and the edge of aluminium-lined seal, defined as mushroom height, was measured using an ABSOLUTE digimatic indicator (model ID0C1012CE, Mitutoyu Corp. USA, Aurora, IL). The liquid crystal display (LCD) display unit can accurately measure linear dimension down to 0.01 mm. The reading was obtained by inserting the mushroom side of the cartridge into a custom-made holder that was connected to the spindle of the indicator. The indicator was previously zeroed (calibrated) by insertion of a flat-topped, cartridge-like stainless steel rod into the holder. In this study, the crimping conditions specified above resulted in a mushroom height of approximately 0.8 mm.
Residual Seal Force (RSF) Measurement
The residual seal force (RSF) of the septum lined seal was determined by a Residual Seal Force Tester (Genesis Packaging Technologies, Exton, PA) where strain (compression) was applied at a fixed rate to the septum-lined seal and the strain (distance) vs stress (force) data were collected. The data were then analyzed by a proprietary data analysis algorithm to determine the RSF. The RSF readings are considered objective (no operator subjectivity is involved in the measuring process) and consistent within ±20%.
Packaging Preparation
The aluminum foil and Tyvek® sheets were cut to the size of, and heat sealed to, a PETG tray. This heat sealing process was performed with custom tooling that fits the shape of the blister pack (Figure 1b). The tooling was designed to apply even pressure and heat to the lid while holding the blister tray flat. The parameters for sealing included temperature, pressure, and duration. In this study, the parameters were set to 260 °F, 80 psig, and 3.7 s, respectively, for the Tyvek® sheet, and 320 °F, 80 psig, and 5 s for the aluminium foil sheet.
Ozone Exposure
A custom-made, stainless steel ozone chamber (Figure 2) was configured with an ozone generation system consisting of a generator (HTC-500S), a relay system (R-25), and a monitor (106-L), all acquired from Ozone Solutions. Cartridges were stored inside the chamber and exposed to various concentrations of ozone established according to parameters in Table I. Samples can be viewed through the viewing window and accessed from the side door. An electric fan (59 cubic feet per minute) was placed inside the chamber (center of the ceiling) to increase air turbulence and enhance ozone uniformity. A pressure gauge was installed to monitor the overall chamber pressure as a safety precaution. The rate of ozone production was controlled by adjusting the dial setting on the O3 generator as well as the rate of air flow.
Ozone Diffusion Measurement
Diffusion experiments were performed to determine the diffusion rates of ozone with the purpose of estimating ozone's diffusion coefficient (diffusivity) across permeable packaging barriers. A hand-held ozone monitor (Aeroqual Series 500 with OZG sensor, Ozone Solutions) was calibrated and placed inside a Tyvek® pouch (9.75 × 7.25 inch, 0.0066 inch in thickness, 1120 mL in total volume, P/N 027710C, Amcor), and the pouch was placed inside the ozone chamber. The hand-held monitor recorded the increase in ozone concentration in the pouch as it diffused in from the ozone chamber where the ozone concentration was controlled at 1, 5, and 10 ppm, respectively (n = 3 for each concentration). The pouch was placed inside a large cardboard box (similar configuration to that in Figure 1c) to simulate ozone's diffusion mechanism into the packaged blister pack.
With data from the diffusion experiments, the diffusion coefficient (D) of ozone could be calculated based on Fickian diffusion. The following model was used to represent the system:
The interior concentration (CI) is modelled to change with time (t). The model is simplified to use steady state flux, since the Tyvek barrier is very thin and relatively easy to permeate. The change in CI is linearly proportional to the surface area of the barrier (e.g., Tyvek® pouch) (A), and the concentration difference (CO − CI) across the barrier. It is inversely proportional to the volume (V) of and the thickness (L) of the package barrier. With the exterior concentration, CO (i.e., the ozone concentration in the chamber) being constant and the initial concentration inside package barrier being zero, eq 1 can be solved and expressed as eq 2:
In this form, the diffusion coefficient of ozone across the permeable package barrier can be calculated from the ratio of ozone concentrations inside and outside the barrier at a given time point. The diffusion coefficient has units of length2/time.
Oxygen Transmission Rate Measurement
Permeability of the molded PETG tray of the blister pack was measured by the oxygen transmission rate per the ASTM F-1307 standard method using an oxygen transmission instrument (Oxtran 2/21L, Mocon Testing Service, Minneapolis, MN). Testing was performed using 100% oxygen under 23 °C/50% relative humidity (RH). Ozone is expected to diffuse slower than oxygen given the larger molecular size.
Optical Microscopy
All photographs for image analysis were taken on an optical microscope (model Vertex 220, Micro-Vu Corporation, Windsor, CA) at 30× magnification with the ring light set to 12% intensity. The microscope station was interfaced with Micro-Vu InSpec metrology software for scanning and graphic analysis. Consistent lighting was essential to the image analysis, as it relied on the difference in brightness between cracked and uncracked surface area.
ImageJ Analysis
ImageJ is a public domain, open source, Java-based image analysis program (http://rsbweb.nih.gov/ij/docs/intro.html) consisting of a number of standard image analysis tools and custom features. Here it was used to analyze images of cracked septum. It was performed by using a color thresholding tool in hue/saturation/brightness space to omit all excess image data, leaving only the cracks visible. In this case, the lower limit of saturation level (a measure of colorfulness) threshold was set to 70, and the upper limit of brightness level (a measure of how much light is being emitted or reflected) was set to 100. With these thresholds established, the image was converted to a binary form (black and white). The particle analysis tool was applied to measure the area of the cracks isolated, treating each crack image as a particle. With the option of specifying the form of particles such as size, circularity, etc., the total crack area (in square millimetres [mm2]) can be calculated. This data can be tabulated in a program such as Excel or JMP.
Results and Discussion
Ozone Chamber Performance
This ozone control system yielded minimal oscillation in ozone concentration inside the chamber (Figure 3). The high and low limits of oscillation were dictated by several parameters (Table I), including the pressure of the in-house air (at 2 psig throughout this study), ozone generator setting (scale of 0–10), monitor setting on target ozone concentration, and data logging frequency.
These parameters were used to set six ozone concentrations, targeted at 0.02, 0.2, 1, 5, 10, and 50 ppm (Table I). Their corresponding ozone outputs, high/low range, and averaged ozone concentration are summarized in Table I. An electric fan (with flow rate set at 59 cubic feet per minute [CFM]) was used to ensure homogenous distribution of ozone in the entire chamber. It was also noted that, when the fan was turned off, ozone concentration fluctuated in a range much wider than in the fan-on condition. For the example of 1 ppm (Table I), the range and average of ozone concentration was 1.0–1.3 ppm and 1.2 ppm, respectively, when the fan was turned on; it was 0.95–2.3 ppm and 1.3 ppm, respectively, when the fan was off. The effect of air movement will be discussed in a later section.
Effect of Ozone Concentration on Septum Mushroom Cracking
Ozone was found to play a critical role in septum cracking (Figure 4a–e). After 8 h exposure to a low ozone concentration (0.02 ppm), the septum mushroom of the naked (not packaged) cartridge began to show subtle cracks only detectable under microscopy (Figure 4b). Cracking was much more significant 4 weeks after (Figure 4c). At a higher concentration (1.1 ppm), ozone attack on the naked cartridge was relentless, showing substantial cracks after only 6 h exposure (Figure 4d). Figure 4e represents a cartridge in a blister pack sealed with a permeable Tyvek® lid. Despite this barrier the mushroom still cracked substantially after 6 h at 1.1 ppm ozone exposure. Interestingly, the center of the mushroom was intact, attributed to the orientation of the cartridge in the blister pack. The cartridge was facing down so the tip of the mushroom touched the PETG trays and was insulated from ozone access. This phenomenon further substantiated that septum mushroom cracking was caused by ozone attack.
The septum mushroom's propensity to cracking could be estimated by how fast the mushroom displays the first sign of cracking observed under microscopy at 30× magnification (e.g., Figure 4b). This method was used to assess the effect of ozone concentration on mushroom cracking as summarized in Table II. For naked cartridges (not packaged), the first sign of cracking on the mushroom was observed 8 h after exposure to 0.025 ppm ozone concentration, but cracks showed up faster with increasing ozone concentration. For instance, the mushroom cracked almost instantly (within 1 min) when the ozone concentration was raised to 50 ppm. This duration-ozone concentration relationship, duration = 17.6 [ozone conc]−0.90, established by fitting the experimental data (R2 = 0.999) from 0.025 to 50 ppm (Table II), allowed the duration for mushroom cracking (the naked cartridge) to be extrapolated to extremely low ozone concentrations. It estimated that the mushroom might crack in 6 ± 2 days at a standard indoor ozone concentration, for example, 0.001 ppm (1 ppb). It appears to be very fast but note that it is under a highly turbulent condition. Again, the effect of air movement on cracking will be discussed later in details.
Effect of Stress on Mushroom Cracking
The stressed rubber is known to be particularly vulnerable to ozone attack, and resulting cracks are always perpendicular to the applied stress (8). In view of the random pattern of the cracks shown in Figures 4b–e, it suggested that the residual seal stress imposed on the mushroom is uniform across all directions. Because the aluminum-lined seal stretched the septum from the circular edge, the top of the septum mushroom was stretched most and developed wider and deeper cracks, as shown in Figures 4c and 4d. For the cartridge in the blister pack with the septum mushroom facing down and touching the plastic tray, a deep continuous crack encircling the central portion of the mushroom formed. In this case the crack developed in the direction perpendicular to that of applied stress, which is consistent with the theory.
Although the septum mushroom is sensitive to the environment of high-concentration ozone (cracking in 1 min at 50 ppm in Table II), the septum disk without any residual seal force (i.e., not crimped) was inert to ozone attack even after 80 h exposure to 50 ppm. Apparently stress serves as a catalyst in crack formation where ozone is a strong reactant. Thus, in theory cracking should stop progressing after the stress on the rubber is released. This theory was tested by exposing the naked cartridges and cartridges in blister pack to 1.2 ppm ozone. At different time points of exposure, pictures of the mushroom were taken and each digital image was analyzed for total crack area in ImageJ. This was a quantitative method better than that of measuring the width or length of a single crack on the mushroom. Table III summarizes the total crack area (in mm2) of naked cartridges exposed to 1.2 ppm ozone for different durations (up to 134 h). The crack area grew initially and then levelled off, suggesting that cracks stopped growing or developing as the stress was released during crack formation.
RSF in the stopper/seal combination is often used to quantify seal tightness as a result of the sealing process. In this case, mushroom height should correlate with seal tightness, thereby the RSF of the septum lined seal. Cartridges with two mushroom heights, 0.31 ± 0.04 (compression pressure of 2.5 bar and a knife height of 5 mm) and 0.82 ± 0.04 mm (compression pressure of 4 bar and a knife height of 5 mm) (n = 4 for each height), were prepared and investigated for ozone exposure. First of all, the RSF of these mushrooms was determined, using a Residual Seal Force Tester, to be 5.0 ± 1 Newton and 8.5 ± 1.6 Newton for mushroom heights of 0.31 mm and 0.82 mm, respectively. The effect of RSF (stress) on mushroom cracking was assessed by exposing these cartridges to 1.2 ppm ozone concentration. The correlation between mushroom height and cracking is obvious; cracking was observed on high mushroom (0.82 mm) cartridges in 9.5 ± 1.7 min (three separate measurements) while the low mushroom (0.31 mm) cartridges did not crack until 31.3 ± 4.4 min (three separate measurements). As expected, the total crack area is larger for the higher mushrooms than that for the lower mushrooms, 0.21 ± 0.03 mm2 vs 0.054 ± 0.044 mm2 after 6 h exposure. At this point, all the data suggest that stress is a critical factor in causing mushroom cracking.
Effect of Packaging Barrier on Mushroom Cracking and the Protective Mechanisms
Two packaging configurations were tested in this study, blister pack (Figure 1b) and blister pack/carton combination (Figure 1c). The blister pack consisted of a PETG plastic tray sealed with a lid membrane. The PETG tray was deemed impermeable based on its oxygen transmission rate of 0.0073 ± 0.0003 cc/package/day, measured per the method of ASTM F-1307 under the condition of 23 °C/50% RH. The lid membrane could be either a permeable Tyvek® lid (Gurley porosity of 70 s/100 cm3 air) or an impermeable aluminium foil lid (oxygen transmission rate of <0.002 cc/m2/day). Using an impermeable barrier (aluminium foil) to completely block ozone access is certainly the most effective solution, which was confirmed by the intact mushroom in this blister configuration after exposure to 50 ppm ozone concentration for 80 h (Table II). However, it is interesting to find that the blister pack with a highly permeable Tyvek® lid could delay mushroom cracking by 8 to 12 fold compared with the unpackaged cartridge (Table II). Upon exposure to the standard indoor ozone concentration (≤2 ppb), the mushroom in the blister pack was estimated to crack after 73 days vs 6 days for the naked cartridge.
Gurley porosity is often used to quantify air resistance for paper-type barriers by measuring the number of seconds required for 100 cm3 of air to pass through 1.0 in2 of a given barrier material at a pressure differential of 0.188 psi. This particular Tyvek® lid was reported to have 70 Gurley seconds (vendor data sheet). Considering the blister pack having an area of 12 cm2 of Tyvek® lid and a volume of 10.8 cm3 in the blister, it may take less than 5 s for the ambient air to fill the blister volume. Certainly, in the real situation there is no positive pressure to drive the air across the Tyvek® sheet and ozone would transport mostly by diffusion.
To investigate whether the delay of mushroom cracking in blister pack was primarily due to the lag time of ozone to reach the external ozone concentration in the blister, we experimentally determined the diffusion rate and used this data to calculate the diffusion coefficient of ozone across the permeable package barriers, Tyvek® sheet, and cardboard carton. Because the commercially available ozone monitor substantially outsized the blister pack (Figure 2b) and the cardboard carton (Figure 2c), a larger Tyvek® pouch and a bigger cardboard carton were used instead. The hand-held monitor was heat-sealed in a pouch made of the same Tyvek® sheet material and placed inside the ozone chamber to generate the ozone concentration profile (i.e., CI as a function of time). With measured CI, ozone's diffusion coefficient was calculated using eq 2. At three chamber concentrations (CO), 1, 5, 10 ppm, the times for ozone inside the barrier to reach 95% of the exterior concentration were plotted in Figure 5. Ozone diffusion rates across the Tyvek® membrane are ozone concentration-dependent: the higher chamber concentration, the faster the diffusion rate. It took 94, 23, and 5 min for ozone concentration inside the Tyvek® membrane to reach 95% of the chamber concentration at 1, 5, and 10 ppm, respectively. The calculated ozone diffusion coefficients across the Tyvek® membrane at these three ozone concentrations as listed in Table IV also confirmed this observation. Such a concentration dependence of gases diffusing across polymeric membranes was reported previously (9). As reported in Table II, cracks developed in 10, 4, and 2 min when the naked cartridge was exposed to these three ozone concentrations. If ozone diffusion across the blister barrier is the only factor that delayed crack occurrence, it could be estimated conservatively that cracks should initiate approximately in 104, 27, and 7 min when cartridges in the blister pack were exposed to 1, 5, and 10 ppm ozone concentrations, respectively. Actually, Table II showed that it took longer: 132, 38, and 23 min, respectively.
The same diffusion experiments were performed to assess the secondary package, a cardboard carton. The hand-held monitor was placed inside the carton directly or after being sealed in the Tyvek® pouch. Upon 5 ppm ozone exposure, the calculated diffusion coefficient is 70.3 × 10−6 (Dcarton) and 3.01 × 10−6 cm2/sec (DTyvek+carton) for the cardboard carton and for the blister/carton combination, respectively (Table IV). The carton provided resistance to ozone diffusion but at a lesser extent than the Tyvek® membrane (DTyvek = 19.9 × 10−6 cm2/s). When the Tyvek® barrier and the carton were combined, its diffusion resistance (1/Dcarton) is more significant than the resistance by each barrier combined (1/DTyvek + 1/Dcarton). All these results suggest that factors other than diffusion may be involved in delaying mushroom cracking.
Effect of Air Flow/Movement on Mushroom Cracking
As effective mixing facilitates chemical reactions or the drying process, air flow may be an important parameter affecting ozone/stress-mediated rubber cracking. Strong air movement (or turbulent flow) would enhance ozone attack by establishing a thin boundary layer with a sharp ozone concentration gradient, creating little resistance to any chemical reactions. Air flow inside the ozone chamber (∼3 cubic feet in volume) was highly turbulent due to the action of a fan (59 CFM). To demonstrate the effect of air movement, ozone exposure experiments were performed on naked cartridges under three conditions: fan on, fan off, and fan-on with cartridges covered by a 50 cc glass beaker on a perforated shelf. As discussed in the section on ozone chamber performance, under the fan-off condition ozone concentration fluctuated in a broader range (0.95–2.3 ppm) and had a higher average ozone concentration (1.3 ± 0.36 ppm) compared to the fan-on condition (range of 1.0–1.3 ppm and average of 1.2 ± 0.14 ppm). Cartridges placed on a perforated shelf (in the middle of the chamber) and covered by a standard 50 cc glass beaker simulated the condition that the cartridges were sheltered and insulated from the direct attack of turbulent air. The openings (holes) of the shelf allowed the ozone concentration inside the beaker to equilibrate quickly. Each group of cartridges (n = 3) was exposed to ozone for 20 and 35 min in the chamber under the same setting, 1.2 ppm (fan on) or 1.3 ppm (fan off). After exposure, digital pictures of the cracked mushrooms were taken and the total crack area was measured by ImageJ analysis (Table V).
Despite the higher averaged ozone concentration for the fan-off condition, the extent of mushroom cracking, as quantified by the total crack area, was much less significant than the fan-on condition at both the 20 and 35 min time points. Cracking was particularly mild at the early time point (20 min) when the air was stagnant (fan off), having a crack area only 25% of that resulting from the turbulent condition (fan on). The same trend was observed for the cartridges covered by the glass beaker under the fan-on condition. Their total crack area was similar to that under the fan-off condition, suggesting that the glass beaker indeed shielded the “turbulent” ozone from attacking the stressed mushroom even though the delay by the diffusion into the beaker is relatively short.
Together, diffusion across and reduction in air movement inside the permeable barrier delay ozone attack on the stressed mushroom. The effect of these two factors may multiply when two permeable barriers are combined. This may explain the much longer time, 145 h (Table II), needed to crack the septum mushroom for cartridges packaged in the blister/carton configuration compared to 24 h for cartridges packaged only in the blister pack. Overall, the final package of these two permeable barriers is estimated to be capable of protecting septum mushroom cracking from occurring for a long period of time, perhaps years, if the product is stored under a standard indoor condition (≤2 ppb ozone concentration and low air movement).
Conclusions
In this study we verified that the root cause of septum mushroom cracking is ozone attack. A standard outdoor ozone concentration (10–40 ppb) could crack the unprotected septum mushroom in days. The time for crack occurrence was shortened and the severity of cracking (in terms of total crack area) was aggravated by increasing ozone concentration. Another necessary condition is the residual stress on the septum, physically cracking the rubber after it is weakened by the chemical reaction between ozone and rubber's diene double bonds. The septum did not crack if no stress was applied. Cracking stopped progressing perhaps due to release of stress by the crack, and the mushroom ended up with similar total crack area regardless of how the cartridge was packaged. Although packaging involving impermeable barriers is the most effective solution to ozone attack, even permeable barriers can delay cracks from developing by reducing convective flow and slowing diffusion. Therefore, combination of the packaging involving two permeable barriers (blister pack/carton) can still protect the septum mushroom from ozone attack during product shelf life if the storage environment maintains a standard indoor ozone concentration.
Conflict of Interest Declaration
The author(s) declare that they have no competing interests.
Acknowledgments
The authors are indebted to Dr. Sherry Martin-Moe for project support, Mr. Jacek Guzowski for ozone chamber fabrication, Dr. Philippe Lam for assistance in lined seals photography, Ms. Nicole Quesada for assistance in lined seal crimping and residual seal force analysis, and Ms. Lilian Liu for sourcing the aluminium foil material.
- © PDA, Inc. 2011
PDA members receive access to all articles published in the current year and previous volume year. Institutional subscribers received access to all content. Log in below to receive access to this article if you are either of these.
If you are neither or you are a PDA member trying to access an article outside of your membership license, then you must purchase access to this article (below). If you do not have a username or password for JPST, you will be required to create an account prior to purchasing.
Full issue PDFs are for PDA members only.
Note to pda.org users
The PDA and PDA bookstore websites (www.pda.org and www.pda.org/bookstore) are separate websites from the PDA JPST website. When you first join PDA, your initial UserID and Password are sent to HighWirePress to create your PDA JPST account. Subsequent UserrID and Password changes required at the PDA websites will not pass on to PDA JPST and vice versa. If you forget your PDA JPST UserID and/or Password, you can request help to retrieve UserID and reset Password below.