Breaking the habits!

How to Prevent Breakage of Glass Primary Packaging Containers in the Pharmaceutical Filling Line

The world is evolving ever faster and so are standards, requirements and developments in the pharmaceutical packaging industry. When the standard filling speed for parenterals was around 100 to maximum 400 pieces (syringes, cartridges, vials or ampoules) per hour this has changed drastically with high speed filling lines to numbers higher than 600 pieces per hour. But this came along with an increased sensitivity for breakage in these high speed filling lines. What sense does it make to save time in filling if you spent it again with cleaning up the splinters from broken containers? So along with the high speed filling lines came the awareness for the quality of the primary packaging containers. But to avoid breakage in the filling line first the fundamentals of fracture mechanics have to be understood.
Fundamentals of Fracture Mechanics

Strength is defined as the capability of a material to develop resistance against external pressure. In order to describe the strength of an object material constants like Young’s modulus and Poisson’s ratio can be used. This is, however, only half the truth. Values like stiffness, toughness, elasticity, plasticity, brittleness, ductility, etc. which are strongly dependent on the geometrical dimensions of the object are also used to describe the strength of a material. For an exhaustive description and a full understanding of glass strength surely all of these properties have to be considered. Some of these influencers are explained in the following article. 

In order to deform (to change the original shape of) a material an external force (a load) has to be applied. This force (F) is acting on a specific area (A) on the material. The material naturally tries to resist this force which creates stress (σ).

Stress (σ) can be defined as the immediate internal reaction (resistance) of a material to an applied external force. In case of a force stretching the material (external tensile stress), the atoms in the glass network would be “torn apart” and they would try to counteract the force by keeping together. On the other side if a force is squeezing the glass (an external compressive stress) the atoms would be pushed together and they would counteract the force by pushing apart (fig. 1).
Fig. 1 Illustration of stress within the glass structure. Left: No stress in the glass. Middle: Tensile stress. Right: Compressive stress.
If stress is put on a material the immediate reaction is a geometrical deformation. So the material is e.g. either bending, caving in or twisting. This deformation is measured as strain. The higher the strain the higher is the deformation. The deformation again can be classified into elastic and plastic deformation. When an external force is applied and removed again the material tries to return to its initial shape. A good example for this behavior is a balloon that regains its small size as soon as the air is released (fig. 2 left). If the external force continues to act, the material can have the ability to remain in the deformed shape, even after the force finally has been removed. An example for this would be metal rods that remain in the deformed shape (fig. 2 right).
Fig. 2 Left: Balloon showing elastic deformation when the air is released. Right: Metal rods exhibiting plastic deformation.
Now we want to put all of the above mentioned parameters together: If stress is applied on a material it reacts with strain, which first means an elastic deformation followed by a plastic deformation before it finally ends up in breakage. This is visualized in figure 3:
Fig. 3 Stress-strain diagram comparing brittle and ductile materials.
If a “small” stress is posed on a material and removed again the atoms in the structure are misplaced but are able to return to their original places again. Depending on the amount of stress both brittle and ductile materials are able to do this. If the stress is continuing ductile materials are starting to permanently deform plastically (see metals rods). Glass as a brittle material is not able to plastically deform. Initially there is a very small range of elastic deformation. However, if the stress continues glass is not able to compensate this by deforming plastically but instead it immediately breaks.
Flaws on the glass surface

Littleton once said about glass strength measurements: ”We do not measure the actual strength of the glass but the weakness of the surface.” [1, p. 365]. With these words he meant that the influence of the surface defects on the strength are actually much higher than the composition of the glass material. 

This understanding required a long series of experiments. It had to be shown that the surfaces of glass do contain flaws. Then the origination of the failure from these surface flaws had to be proven. And finally, the reconstruction of the propagation of the cracks had to be done in order to being able to predict the lifetime of that certain investigated glass piece.
Pioneers in these fields were C. E. Inglis, F. W. Preston and A. A. Griffith in the beginning of the 20th century [2] [3] [4] [5] .

Surface flaws can be generated throughout the whole handling process from the tubing through the converting process up the filling and final packing. Each and every one of these flaws can contribute to a decrease in strength, i.e. break force and increases the risk for breakage.
Fig. 4 Schematic overview of the value chain of a glass primary packaging material; showing that from tubing manufacturing and converting and throughout the whole pharmaceutical filling process the strength of the glass (the break force) is decreased by the continuous addition of surface defects.
These flaws do not only refer to defects on the glass surface, like scratches and cracks but also to stress inside the glass induced by improper annealing (in the converting process) or heat treatment in general, like depyrogenation tunnel, freezing and thawing and lyophilization (fig 5).
Fig. 5 Pictures of micro-cracks, hit spots and scratches (indicated by a blue arrow) on glass container surfaces.
Final breakage of the glass container

The probability of breakage of a brittle material like glass is not correlated to a single material property but rather to an interaction of surface flaws and mechanical stress (tensile stress created by a mechanical or thermal impact).

Thus, breakage of glass is a function of both the quality of the glass surface and the mechanical tensile stresses that the glass is exposed to. These criteria for breakage of brittle materials are expressed by the so called Griffith equation:

KI = σ∙Y∙√c 

σ = stress (induced by an applied force)
Y = geometrical factor (which considers, among others, the location of the defect)
√c = critical dimension (e.g. depth of defect)

Hereby the stress intensity factor KI is expressed as the product of an externally applied tensile stress (σ) and the shape/geometry and the size of a flaw (expressed by Y and c, respectively).

Now, if the stress intensity factor KI reaches a critical, material-dependent value, breakage of the material will occur. The critical value at which breakage will occur is called fracture toughness. The fracture toughness is a material-depending constant whose values range between approximately 0.6 and 1 MPa√m for glass.

So the criteria for breakage of glass depend on the interaction of the size and shape of the flaws present on a glass surface and the extent of mechanical tensile stresses applied to these flaws.

Then, the value of the mechanical tensile stress at which the breakage criterion (fracture toughness) is reached or exceeded first, is defined as the strength of the material. So glass can fail under low applied mechanical stresses (i.e. exhibiting a low strength) when the size (c) and the shape (Y) of a flaw exhibit high values (fig. 36 left). On the other hand, if the flaws exhibit low values for c and Y, high mechanical stresses (i.e. exhibiting a high strength) can be applied until the critical stress intensity is reached (fig. 6).
Fig. 6 Fracture toughness being exceeded by Left: a significant surface defect and a small applied stress, Right: a small surface defect and a significant applied stress.
Thus, a very important fact now becomes clear: The strength of a borosilicate glass container does not depend on the glass composition. The predominant criteria are indeed the pre-damages and final stress intensity factors.

How to prevent glass containers from breaking

Now, the most important point to prevent breakage is to prevent the formation of defects on the glass surface in the first place. Second, the stress which is applied on the glass should be minimized as much as possible. A number of possible sources for the creation of surface defects are listed below:

Possible sources for surface defects can be (among others):
- Scratches from transportation
- Cosmetic defects inside the glass (e.g. stones, knots, inclusions)
- Scratches from glass-to-glass contact
- Scratches from glass-to-metal contact
- Thermal shock micro-cracks from hot glass touching cold material
- Scratches from transportation belts
- Scratches from guide bars in the filling line
- Residual thermal stress due to improper annealing
- Scratches due to defective forming tools
- Scratches due to transfer arms

Possible external forces triggering the breakage can be (among others):
- Glasses hitting each other
- Glasses hitting a metal guide bar
- Glasses hitting the metal edge between two modules
- Improperly adjusted crimping or capping tools
- Volume expansion in lyophilization process

A targeted risk assessment should include the systematic evaluation of each step in the whole process for potential introduction of surface defects and improper handling. Metal should be used as little as possible as a contact material and the machines should be precisely adjusted to the dimensions of the containers.

Since in daily business it is impossible to determine the size c and shape Y of all flaws present on a glass container, the strength (σ) of a whole batch of glass containers cannot be predicted. Due to the difficulty of measuring the size (depth) and the three-dimensional geometry of a flaw by means of optical inspection tools, there is no existing non-destructive inline method available for the determination of the actual strength of each glass container up to now. Optical inspection systems normally only record the superficial dimensions of defects whereas the depth of the defect is not identified.
In other words, optical inspection systems are capable of detecting cosmetic defects, but their capability of interpreting and evaluating flaws in terms of criticality for strength is very limited.

So the only strategy for a reliable assessment of the strength of glass containers is appropriate strength experiments. Appropriate strength experiments enable an objective, quantitative tool for the assessment of the strength of glass containers and should be an essential tool for a comprehensive risk management of a pharmaceutical company.


1. W. Vogel, Glass Chemistry, 2 ed., Berlin: Springer, 1994.
C. E. Inglis, „Stresses in a plate due to the presence of cracks and sharp corners,“ Transactions of the Royal Institute of Naval Architectes, Bd. 60, pp. 219-241, 1913.
3. F. Preston, "The Structure of Abraded Glass Surfaces," Trans. Optical Soc., vol. 23, pp. 141-64, 1921.
4. F. W. Preston, "The Mechanical Properties of Glass," J. Appl. Phys., vol. 13, pp. 623 - 34, 1942.
A. Griffith, "The Phenomena of Rupture and Flow in Solids," Phil. Trans. R. Soc. Lond. , vol. 221, pp. 163-198, 1921