Published in International Pharmaceutical Industry, Autumn 2018 Volume 10 Issue 3: http://ipimediaworld.com/product-load-effect-thermal-properties-product-affect-package-performance/
By Jerry Ferracamo, Product Development Manager, Inmark
“There’s a way to do it better – find it.” – Thomas Edison
For those of us familiar with cold chain packaging design, we focus much of our attention (and rightly so) on insulation and refrigerants as the primary tools in our toolbox. We creatively endeavour to balance resistance and heat capacity to keep a desired volume within acceptable temperature limits, using tricks of the trade (thermal breaks, free convection, phase change, etc) to meet even the most challenging requirements. While at times thermal design may seem akin to mastery of the dark arts, it is really a function of how we manage the interaction of variables based in the laws of heat transfer. While it would be more fun if all it took was an eye of newt, some witch’s bane and a hair from some mythical beast tossed in a cauldron, the (somewhat more boring) reality is that understanding the physics behind conduction, convection, and phase change are the key ingredients in our brew.
One variable that can have a significant impact on system performance that doesn’t get much attention is the nature of the product to be kept at temperature. From a designer/engineer’s perspective, we know that a wide variety of products, in different form factors, primary packaging, bulk packaging, etc, will likely be shipped in our designs. How we arrive at what we ultimately “qualify” in our laboratory studies is largely a function of whatever water-filled bottles or vials we have lying around, and if we have enough of them for our n=3 tests. I guess our carefully-crafted brew might be more of a potluck. Or more nefariously, it is also possible to “engineer” a pass in a system design by using product load properties to advantage (hello block of lead!).
For those of us who use packaging to move temperature-sensitive goods, especially today’s advanced medicines such as biologics, the question is: Will this package work for my product? Short of fending off threat of temperature excursion with a Patronus charm, a little bit of mastery in the dark arts of physics can help pharma professionals ask the right questions and make better decisions on the packaging that it employs to move its life-saving therapies.
As many of us know, ambient temperature profiles are a significant determining factor in how packaging designs are evaluated. Profile generation and selection are certainly dark arts as well (that we’ll cover in another article), but the underlying goal of selecting the appropriate ambient temperature profile(s) generally falls into one of two camps: representative or worst-case. The camp one falls in is generally dictated by the risk profile of the product(s) to be transported. So for example, if Biopharmaceutical Company A manufactures a biologic that takes 6+ months to produce and the product retails for $1500/vial, the risk profile for their drug might look very different from Pharmaceutical Company B, whose more traditional pharma drug can be produced in weeks and retails for $100/vial. As such, Company A might opt for worst-case profiles, which could bound ambient exposure for >99% of their shipments. Company B, however, might want to consider more representative profiles for their lanes. Why? By accepting some amount of risk of temperature excursion, Company B can hold down packaging and transportation costs by utilising the 80% solution as opposed to the 99%+ solution.
Like with ambient temperature profiles, much of the prevailing industry approach regarding product load involves a bracketing scheme (mass-based), in which a “light” minimum load and a “heavy” maximum load are qualified, and the logic (incorrectly) applied is that any product that weighs between the “light” and “heavy” payloads will be bound by the results from the two loads qualified. Ah, if only the world were so simple…
Our problem here is not one-dimensional. Let’s say, for this discussion, that our payload volume is a well-defined space (i.e. product carton). The temperatures within that volume will vary throughout the volume, and the temperatures and amount of variation directly results from what’s in the box and its interaction with the packaging design. So our problem then must consider the thermal properties of what’s in the box, as well as where in space the product resides.
So if we think about “what”, some of the thermal characteristics at play are:
- Heat Capacity (cp): the amount of heat required to raise a system’s (product load’s) temperature by 1°
- Thermal Conductivity (λ): the quantity of heat transmitted through a unit thickness of a material
- Thermal Diffusivity (κ = λ/ρcp): represents the ratio between a material’s thermal conductivity to its heat capacity and density
In terms of “where”, that is simply how much of the given volume the product occupies. It goes without saying that the further to the extremities the product pushes into the volume, the more temperature variation is likely. Why? Part of the answer is the thermal characteristics of the product in question. How much heat can it store? How quickly does it conduct heat? Is it discretised (i.e. commercial product) or is it more of a single mass/volume (i.e. carboy, tank, etc)?
Another piece of the puzzle relates to the packaging design itself. Is it susceptible to a wide temperature distribution? Does the design have notable hot or cold spots? A disadvantageous product load could exacerbate the spread in temperature distribution, especially for lower-tech designs that rely on brute force to meet thermal requirements (you want me to put how many gel packs in this box?). The flip side is that an advantageous product load (bring back that block of lead!) can be enough to push a design over the hump to a pass (if it seems too good to be true it probably is…).
In addition, it is also important to consider the role product can play in process-related variables, such as pack time and the temperature equilibration that occurs after packing. Some folks in the industry use the terminology “hockey stick effect”, which is merely a function of the product load cooling after packing, due to its rise in temperature during the packing process. Any thermal packaging design requires time after the package is assembled to equilibrate thermally, as there are materials of various thermal characteristics starting at different temperatures interacting. A product of a low thermal mass will tend to rise in temperature fairly rapidly in a room-temperature environment, then may take some amount of time to cool while the shipping system equilibrates. This phenomenon can cause excursions if not accounted for (generally a start delay on a temperature monitoring device will be used to account for this).
So what does this all look like applied? In this case, we have taken our Răcire 5L design (Figure 1), which is designed as a 48-hour 2-8°C solution, and cycled it against the ISTA 7D 48 summer profile containing four different product loads. Three of the four product loads tested were comprised of 20 x 120mL bottles. The bottles were fully filled for the 1A product load (maximum load used for OQ testing), half-filled for the 2A product load, and empty for the 3A product load. The fourth product load tested was a single, empty 120mL bottle, centred in the product box with bubble wrap (minimum load used for OQ testing). Each of the three full loads was fully instrumented with T-type thermocouples monitoring the front left corner, front face centre, centre, right face centre and back right corner. The single bottle load was instrumented with two T-type thermocouples for redundancy.
Figure 1: Inmark Răcire 5L 2-8°C Temperature-controlled shipping system. The system is comprised of water (solid, shown in blue), PCM4 (liquid, shown in pink), and inner and outer insulation, as well as corrugated shipping and product cartons.
As you can see in Figure 2, we have a plot of spaghetti (yum!). But as it turns out, this plot of spaghetti tells us quite a bit. During the initial packing and equilibration period immediately following packing (start of test), we can see a clear split between the payloads with water-filled bottles (1A, pink and 2A, green), and the payloads without (3A, light blue and 4A, purple). This demonstrates the equilibration period required for product loads with low thermal mass. While this specific design has a short equilibration period, other designs may not. The heavier, water-filled payloads are much more thermally resistant to change, and their thermal inertia prevents these products from a rapid spike in temperature during the short exposure to room temperature during packing. Understanding the characteristics of the product(s) to be shipped will help in defining process controls (ex. maximum time out of refrigeration prior to packing, setting temperature monitor start delays, etc). It is important to keep in mind that the warmer a product load gets, the more heat is being introduced to the packaging system. Differences in starting temperatures of the product load can certainly have an impact on packaging system performance, and can ultimately play a role in whether a test/shipment is a “pass”, or a “fail” with a temperature excursion to disposition.
To that end, if we now focus on the end of the test, we truly see the impact the product load can have on packaging system performance. The durations to failure (exceeding 8°C) were as follows:
- 1A: 55.58 hours
- 2A: 52.83 hours
- 3A: 48.25 hours
- 4A: 49.17 hours
As we would expect, our fully water-filled product load survived the longest, our half water-filled product load was next, and the two empty loads went the shortest durations, failing just under an hour apart. The most striking difference is that between 1A (the fully water-filled product load) and 3A (the full empty product load). These two product loads used the same bottles, same quantities, and the same thermocouple locations. But the difference in duration below 8°C was over seven hours. So think about the cold chain packaging your business utilises. Do you believe your designs were designed with the robustness to accommodate seven-hour changes in performance? Do you see risk based on how your packaging was qualified vs what you actually ship? In a world where a single temperature monitor reading outside of the temperature storage requirements can mean dozens of man-hours spent investigating and dispositioning product, it is imperative to consider all variables in packaging performance, and hopefully as has been demonstrated, product load must be one of the variables considered. It can make all of the difference between a pass and a fail; product delivered to the patient or product delivered to a disposal facility…
Figure 2: Temperature vs Time Plot of Răcire 5L 2-8°C tested with various payloads. The empty product loads (3A, 4A) spike up in temperature initially, eventually fall below the water-filled payloads (1A, 2A), then cross back over the water-filled loads as the systems fully exhaust their heat storage capacity
In summary, understanding how a product load might impact the performance of a packaging design is to understand what the product is, how it’s packed, space relative to the total available space that the product occupies, and the performance characteristics of the packaging design. The idea is not to get so specific that you send out vials and cartons for thermal analysis or spend time researching material properties in your engineering handbook (unless you intend on doing thermal modelling). Instead the idea is to understand the forces at play and to recognise what is representative and what is worst-case, depending on the risk profile of your product, shipment, lane, etc. More importantly, it is to be aware to ask how a given packaging solution was qualified and to determine if the product loads used are effective analogs to your product. At the end of the day, the pharma professionals responsible for management and execution of shipping temperature-sensitive drug product seek to put in place packaging and processes to manage the safe transit of their therapies. Product load can be one of the hidden variables that could cause excursions and/or explain differences between laboratory and in-field performance. While not a panacea, understanding the nature of the product(s) to be shipped is part of the overall puzzle pharma professionals may want to consider as they look to effectively manage risk and put in place the best overall solution(s).