Abstract
Glass delamination has developed as a quality problem for primary packaging containers over the last years. Beside other factors the container production process can contribute to this phenomenon, and it seems mandatory to steer and control the critical parameters. With the Schott Delamination Quicktest, a rapid test method is developed that provides a rough measure for the container-related properties that affect delamination risk in a simple and reliable manner and in a relatively short time span. The application of this testing method minimizes the risk of unexpected corrosion behavior and can be used as a release criterion for the running production.
LAY ABSTRACT: If pharmaceutical products do not match properly with the glass containers used, the creation of delaminated glass flakes can occur. This process is slow and time-dependent and mostly visible after month or years. With the Schott Delamination Quicktest a rapid method is developed that provides a fast measure for a rough indication of a higher risk for a delamination behavior. The application of the test gives quick evidence and can be used as a release criterion.
Introduction
The appearance of glass flakes, so-called lamellae, in pharmaceutical parenteral products stored in primary packaging containers made out of glass came to the attention of both health authorities and pharmaceutical industry within the past years. A variety of interest groups and papers are focused on collecting data and information about the root causes, accelerated lifetime testing, as well as prediction of the container behavior (1⇓⇓⇓–5). Screening of various publications dealing with the delamination phenomenon led to a clear understanding that glass delamination is the result of chemical attack on the interior container surface involving well known glass corrosion mechanisms (6), namely dissolution by hydrolysis and ion exchange (leaching) that depend on the pH (7). The visible outcome in the drug product is the appearance of “glass flakes” that arise in the majority of cases months or years after filling. The chemical properties of the drug/formulation components and their ability to promote glass corrosion are the dominant factors for the generation of these flakes. When delamination occurs, glassy flakes are detached from the interior surface after a storage period, as can be seen in the scanning electron microscopy (SEM) image of a cross-section from an affected area (Figure 1). The visible surface topography consists of detachment zones with a small thickness that will result in lamellae as soon as the corrosive attack destroys the residual connection to the surface.
Inner glass surface of a highly corroded borosilicate glass container (SEM image with a magnification factor of 100,000).
Corrosion of glass by pharmaceutically relevant chemistries (i.e., citrate, phosphate, acetate buffers) and within respective pH regimes (acidic, neutral, alkaline) have been known for decades (8). Beside some obvious parameters—like glass composition, storage temperature, time, and other environment conditions—a variety of other factors from container production, subsequent chemical treatment, and processing at the pharmaceutical filling line can influence the delamination process. If delamination is observed in tubular converted glass containers like vials, it starts at the areas where heat was applied during the forming steps. In most cases it can be found near the bottom; sometimes it is near the neck. These areas are reheated to temperature regimes that allow a new arrangement of local chemical bindings by fast physical and chemical mechanisms like evaporation, condensation, and diffusion. By the end of this process, the near-surface glass composition can be altered and a glass phase separation into a silica-enriched and a borate-enriched phase can be initiated (9). After filling the drug product, chemical interaction processes occur and initiate corrosion mechanisms at the whole glass surface. Especially the mentioned vial areas close to bottom and neck might feature a lower chemical durability and therefore a higher vulnerability in comparison to the not-reheated glass surface of the vial body region. Due to the fact that the delamination phenomenon is always a result of an individual corrosion mechanism between glass container and pharmaceutical drugs, a reproducible quality level of the inner surface of the glass containers seems obvious to reduce the entire complexity. In a first step the interaction mechanisms between the individual pharmaceutical product and glass container is checked by predictive screening methods (10) similar to procedures described in USP <1660> or (3⇓–5) to assess the delamination risk. But for quality assurance also a Quicktest method is necessary to control the corrosion behavior of each individual lot comparable to the procedure for the test of the hydrolytic resistance of the interior surfaces of glass containers according ISO 4802 or Ph. Eur. 6.0 Chapter 3.2.1. In this paper we describe the application of a test method, filed as a patent (11), which allows a screening of the general corrosion behavior within a short time interval (less than 1 day).
Materials and Methods
Materials
Vials of different sizes (nominal filling volume between 2 and 20 mL) were produced on state-of-the-art converting lines using glass tubes made of FIOLAX® (Type I borosilicate glass). After production the vials were tested according ISO 4802 part 2 to check the hydrolytic resistance of the inner vial surface.
In addition, a rapid test method—the so-called Schott Delamination Quicktest—was applied according to the methods detailed in a Schott patent (11).
Methods
Schott Delamination Quicktest.
The test method for evaluating the delamination in glass packaging container, in particular for the pharmaceutical industry, comprises the following steps:
- Step 1: Exposing the glass container to an atmosphere consisting of steam in order to form a corrosion zone (“Smart corrosion process”).
- Step 2: Solving glass components via leaching in ultrapure water and quantifying the concentration of dissolved glass elements as a quantitative method and as an option: Visualizing the corrosion zone by appropriate methods.
Before starting the test procedure the vials are cleaned carefully to remove any debris or dust by rinsing each container carefully three times with hot water (50–60 °C) (with a gentle stream, without pressure and shaking of the vials) followed by a cooling down phase (5–10 min) in which they are filled to the brim with water. Immediately before testing the containers are emptied, rinsed three times with ultra-pure water, and allowed to stand for 5–10 min. Finally the emptied containers are dried with an N2 gas stream.
Step 1: Smart Corrosion Process.
The vials are placed inside an autoclave bottom-up on an appropriate stainless steel net tray to allow permanent exchange with the autoclave's atmosphere. The temperature is increased to 100 °C and then the steam is allowed to issue from the vent cock for 10 min. Then the vent cock is then closed and the temperature is increased to 121 °C with a 1 °C/min rate. Then the temperature is maintained for 240 min at 121 ± 1 °C followed by a cooling down procedure at 0.5 °C/min and venting to prevent vacuum. As soon as the temperature inside the autoclave is below 95 °C, the stressed vials are removed according to common precautions. After cooling down the vials, step 2 is started immediately.
Step 2: Quantitative Testing.
The containers are filled with water R1 up to 50% of the nominal filling volume. In order to enhance the sensitivity of the method, the filling volume is adapted to cover just the potential delamination zone. Each single container is capped loosely with a piece of aluminum foil. After closing the autoclave the procedure described in ISO 4802 is started to heat up at least to 121 °C. Then the temperature is maintained at 121 ± 1 °C for 120 min. The cooling down procedure according ISO 4802 is followed. By using a pipette, the containers are emptied and the extracted filling of all the containers of the same sample set are filled into a pre-cleaned plastic tubes. These plastic tubes are used as feeders for the following analysis step via flame atom absorption spectrometry (FAAS) or inductively coupled plasma optical emission spectroscopy (ICP-OES).
Visual inspection.
To visualize differences in the roughness and the morphology of the inner surface, a colorimetric test is applied to stain the higher corroded areas just by dyeing. Using a solution based on methylene blue, the containers are filled to the brim. After a reaction time of several hours they are emptied and inspected by the naked eye.
As an alternative, stereomicroscopic examination of the heel region of a vial also provides corrosion attack information via interference color banding (10), which will be termed coloration in the following.
Equipment
Autoclave.
A standard autoclave according to the requirements of ISO 4802 is needed, which allows the setting of defined temperature gradients and a temperature accuracy of ±1 °C and a water vapor–saturated atmosphere. The autoclave must be equipped with a tray system that allows the positioning of the vials bottom-up in a perpendicular position. For the testing procedures described in this paper we used a SYSTEC DX 150 autoclave.
Flame Atom Absorption Spectrometry (FAAS).
A state-of-the-art flame atom absorption spectrometer is used to measure sodium oxide up to 8 μg/mL. For our measurements an Agilent FAAS Spectraa 280 FS was employed with appropriate parameters like: 2.0 L/min acetylene, 13.5 L/min air; detection at 589.0 nm with a slit width of 0.5μm and a calibrated detection range of 0 to 8.0 mg/L Na.
Stereo-Microscopy.
For the visual (optical) inspection of the vulnerable area in the heel region of the containers a stereo-microscope (Zeiss SV 11) equipped with a camera system was used.
Secondary Ion Mass Spectrometry (TOF-SIMS).
For ion sputter depth profiling we used an ION-TOF 4 analysis system, equipped with a Gallium analysis gun, an Oxygen sputter gun and an appropriate electron gun to avoid charging-up effects.
Scanning Electron Microscopy (SEM).
Finally a SEM (LEO 1550 including EDS) was used to characterize the morphology of surfaces and cross sections.
Chemicals
Water Quality.
To remove any debris or dust water R is used, which is defined as purified water with conductivity less than 5 μS/cm at 25 °C.
For the filling of the vials according to the employed test methods water R1 is used, which is made out of freshly prepared purified water R and having a conductivity less than 1 μS/cm at 25 °C. In addition CO2 must be removed shortly before analysis by boiling in an already used quartz or borosilicate glass flask/beaker and subsequent cooling to room temperature.
Results
Characterization of the “Delamination Zone”
As can be seen in Figure 2, the inner vial surface is divided into three different areas:
Evaporation area: During bottom forming, the glass in this area is heated up to appropriate forming temperatures in the range of 800 °C and above. Species like Boron and Sodium tend to start evaporation during the bottom forming process. After finishing the bottom forming, a silica enriched bottom glass surface is present.
Condensation area: Evaporated species coming out of area a) condensate at the inner glass surfaces of the container due to the fact that the temperature is far beyond the transformation temperature Tg. This condensation process leads to an inhomogeneous coating of the inner walls with typical evaporation relicts like Sodium borates.
Diffuse-in area: In that area the inner near glass surface is exposed to a high temperature gradient for a short time period (a few seconds) between forming temperature and Tg. Therefore, this geometrical small area features appropriate conditions to allow a diffusion of replaced components into the near surface which generates a change in local glass composition and results in a higher risk for delamination.
Scheme of the inner surface of a vial.
To verify this model we performed SIMS depth profiles to analyze the qualitative elemental depth distribution on both, the diffuse-in or delamination zone (area c) and the condensation zone (area b).
As depicted in Figure 3d, the profile of Boron marked with a dotted ring shows a clear signal enhancement, whereas the other Boron profile (Figure 3b) as well as the profiles of the outer container surfaces (Figures 3a and 3c) indicate a Boron depletion in the region of the analyzed surface.
TOF-SIMS profiles of inner and outer surface regions of a glass vial.
Due to the fact that Boron always is a typical candidate of the glass elements tending to early evaporation during melting or reheating processes, the Boron depth profile of the delamination risk zone indicates a clear hint for a temperature-induced process resulting in Boron diffusion into the glass. This micro-zone exhibits therefore an enriched Boron content which is known in the literature to be prominent for a phase separation from silica-based glass to a sodium borate–based glass phase (12). The argumentation is also supported by the shape of the silicon profile, which clearly features depletion in Figure 3d. For the interpretation of all the depicted SIMS intensity profiles, we have to take into account that the intensity profiles do not directly reflect the chemical composition. The intensity of SIMS signals is extremely dependent on the actual chemical matrix of the secondary ion emission area (13). In addition, there is a difference in the surface composition of the inner and the outer surface caused by the manufacturing process of the tubes (Figures 3a and 3b) that overlaps with the changes generated by the converting process described. Therefore, the most reliable information of Figure 3 comes from a direct comparison of the profiles of outer and inner surfaces (3a with 3c and 3b with 3d): Comparing the profiles of the inner surface, B and Na feature an enhancement in the diffuse-in area (Figure 3d) relative to the condensation area (Figure 3b) and a corresponding signal reduction for Si and Al. Besides these more or less stoichiometrical effects (i.e., if elements are enriched others must be reduced), some additional mechanisms like the known reaction phenomenon of B with Na and Al could play a role too. Especially a slight enhancement of Al2O3 in conjunction with a high temperature enforces a sodium ion transfer from the boron-oxygen network to AlO4 tetrahedron, changing a number of four-coordinated boron into three-coordinated boron (14). Taken into account that the B-O-Al bond formed with three-coordinated boron in Na2O-B2O3-SiO2-Al2O3 glasses has a higher bonding energy, the build-up of a new glass phase can be initiated in comparison to the bulk glass.
To verify the different corrosion behavior, two types of glass vials were produced (one under standard condition, type A, and one under modified condition, type B). Standard condition means correct line settings to produce vials that meet all specifications. Modified conditions are optimized line settings to reduce the condensation phenomena to a minimum in order to minimize the phase separation in the delamination zone (see Figure 2). The lines were operated continuously and samples were drawn on a regular basis (every 8 h). Using the Schott Quicktest, the samples were stressed as described before (step 1). Then a portion of the samples went into step 2 followed by FAAS measurements of the sodium concentration in the solution. The respective values are depicted in Figure 4a over period of 13 weeks. The Na content of type A vials clearly show a higher mean value and a broader distribution over time. Under worst-case conditions higher Na-contents up to 6 μg/mL have been observed during an occupational qualification run that correlates with a coloration classification of 2.
FAAS-results of Na content by performing the Quicktest method.
A higher sodium concentration in the solution is a strong indicator for higher vulnerability of the interior vial wall resulting in higher leaching rates. In addition, the more qualitative method of inspecting the vials with a stereo-microscope in the wall near the bottom area supports the results. Within this area a coloration phenomenon can be seen under special conditions that indicates a reaction zone with modified refraction index after step 1. The samples were rinsed and inspected with the interference coloration method and compared to references. The references are divided into four groups, starting with 0 for no visible coloration up to 3 for strong coloration (see Figure 4b).
Qualitative behavior after sample coloration (each measurement point represents the mean value out of 5 single samples).
Note:
0 = no coloration
1 = slight coloration
2 = moderate coloration
3 = strong coloration
In the case of the vials produced under modified conditions, nearly no coloration is visible. For the standard condition vials, a wider spread is observed, so that a variation between no coloration and moderate coloration can be found.
These results will be strongly supported by a lifetime test under moderate temperature of 60 °C (10) using a storage of both vial types filled with a 15% KCl solution for 6 weeks. This ensures a reasonable acceleration without changing the underlying mechanism of glass attack.
A visual inspection of the heel area via stereo-microscopy indicates a strong corrosion behavior for type A (Figure 5a) and no visible corrosion effects for type B (Figure 5b).
Stereomicroscopy images of Type A vials after 6 weeks storage filled with 15% KCL solution at 60 °C.
Stereomicroscopy images of Type B vials after 6 weeks storage filled with 15% KCl solution at 60 °C.
The first hints of the start of the delamination process are visible on one vial of the type A samples and can be seen in Figure 6.
Stereomicroscopy images of type A vials after 6 weeks storage filled with 15% KCl solution at 60 °C. The arrow indicates the start of delamination.
Conclusion
The SCHOTT Delamination Quicktest was applied to characterize two vial types with a different corrosion behavior. The presented results clearly show that this test method can be used as an indicator for a critical corrosion behavior that can enable glass delamination in a final state. Up to now no testing method is available that can be used as a first indicator for delamination propensity. The Quicktest can be performed within 8 h, so that a sampling during process-controlled vial production can be used as a proof and release criteria for corrosion behavior and an early indicator for higher delamination risk. Even though glass delamination is a drug–container interaction phenomena and therefore strongly dependent on the drug and buffer composition, the principle corrosion behavior of the vials can be monitored via the Quicktest and additional risks can be avoided.
Conflict of Interest Declaration
The authors declare that they do not have any financial or non-financial competing interests related to the content of the manuscript.
- © PDA, Inc. 2014
Jump to section
Related Articles
Cited By...
- Characterization of a Novel Particle in a Pharmaceutical Drug Product
- Comparative Leachable Study of Glass Vials to Demonstrate the Impact of Low Fill Volume
- Historical Review of Glasses Used for Parenteral Packaging
- Comparative Delamination Study to Demonstrate the Impact of Container Quality and Nature of Buffer System
- Particulate Study on NeoProfen, a Neonatal Injectable Product