7 MIN READ

A particle’s perspective: why is it so difficult to create a dry powder aerosol?

David Harris

David Harris

Head of Respiratory Drug Delivery (2012-2016)

The real issue with respirable particles is that in order to be respirable, they have to be really, really small – about a fiftieth of the diameter of a human hair, and this small size presents many challenges in terms of understanding their behaviour – both within an inhaler airway and on their journey into the lungs. Respirable particles typically have an aerodynamic diameter of 1 to 5 μm depending upon which region of the lung they are targeting. “Classical” (read “predictable”) mechanics begins to run out for particles below 2 μm – they become more subject to effects such as Brownian motion (diffusion), electrostatic and other less predictable influencing factors. Because of these factors it becomes much more difficult to understand and control a particle’s likely behaviour.

Another effect of reducing particle size is that the surface-area-to-volume ratio of the material increases. Electrostatics will have a much greater effect on smaller sized particles because of this – and consequently smaller particles will acquire a higher specific charge (charge-to-mass ratio) through contact electrification (triboelectrification) simply due to the higher chance of contact (higher surface area) per unit of mass. And just to complicate things further, in low humidity environments any electrostatic effects are exacerbated due to an increase in surface resistivity through lack of relatively conductive moisture.

particles-divide

it’s extremely difficult to achieve a meaningful simulation of aerosol behaviour within a dry powder inhaler

Even with the most powerful computers available today, it’s extremely difficult to achieve a meaningful simulation of aerosol behaviour within a dry powder inhaler (DPI), simply because the underlying physics is not well understood. In contrast, and with sufficient computing power, the lift and drag coefficients of a passenger aircraft could be predicted to within a few percent, in order to optimise the efficiency of the aircraft design.

To add to all this complexity, respirable API (active pharmaceutical ingredient) particles are usually blended with much larger, inert “carrier” particles of lactose. The physics behind this carrier fraction is on a very different scale to that of the respirable particles – typically the lactose particles are around 10,000 times the mass of the much smaller respirable particles. And almost all blends are at least 95% lactose (w/w), so it’s actually the carrier fraction that dominates the overall behaviour during the transition through the airway of the DPI.

It’s worth noting that the close range adhesive forces that are responsible for holding agglomerates of particles together (Van der Waals, electrostatic and capillary) generally decay with the inverse-square of the separation distance (1/r2). This means that the respirable particles that are attached directly to the surface of the carrier particles have a very strong force of attachment and consequently they are very difficult to separate. Those that are attached via other respirable particles, e.g. in a dendritic structure, are held with a much weaker, net adhesive force, and therefore far more likely to become aerosolised when the inhaler is used.

particles-divide

How much energy is required to separate a drug particle from a carrier?

We know that increasing the separation energy increases the chance of separating (detaching) micronised budesonide API from a typical commercial-grade lactose. But how does this compare to the quantity of energy that is available from an inspiratory manoeuvre?

In 2010, a group of us studied the lung characteristics of 90 healthy individuals (ranging from 4 years old to over 50) and found that the quantity of inspiratory energy available was related monotonically to the individual’s height, and independent of age or gender. Even the shortest subjects in this study consistently achieved approximately 3 J of inspiratory energy – which is several orders of magnitude higher than the quantity required to separate 90% of budesonide particles from lactose.

 
If so much energy is available, why are DPIs inefficient?

One particular challenge with carrier based formulations is that the API particles are a tiny proportion of the blend, and in a well-mixed and homogenous formulation are evenly attached to the carrier particle’s surface. When the user inhales and the airfl ow through the airway of the inhaler entrains the formulation, the API particles do not experience any aerodynamic drag force as the vast majority are effectively shielded in the boundary layer around the carrier fraction.

The physics associated with a single lactose carrier particle as it travels through an inhaler airway is hugely complicated. And it transpires that there is a huge body of scientific research underway looking at a similar scenario of particle interaction – the physics of sandstorms.

 
How do sandstorms reach such huge altitudes?

Sand particles are much larger and heavier than respirable particles, and really not particularly airborne. But sandstorms can reach altitudes of two miles, and travel across entire continents! Only recently has research demonstrated a plausible explanation to these phenomena… It’s all about momentum exchange. For example, if you drop a large bouncy ball with a much smaller second one on top of it, when they hit the ground, the smaller ball will fly off much higher than the height at which they were dropped from. This effect underpins the way in which sandstorms gain altitude – the countless particle-particle impacts result in the smaller ones bouncing higher and higher, and collectively the entire sandstorm can reach huge altitudes. Although there is much similarity in the underlying science, it appears that these two similar but separate research areas have never “collided.”

particles-divide

So how do inhalers actually work?

Let’s face it – we’ve been working on them collectively for six decades, and the market leading device is about 25% efficient – even this seems like an amazing achievement given how hard it is to separate closely bound particles. So what can be done to improve this?

It’s about maximising the chance of deagglomeration… CFD and other techniques enable accurate simulation of high efficiency swirl chambers, so that powerful mathematical models can then be constructed from the CFD data to allow interpolation of the design space. Proxies for deagglomeration efficiency can be estimated for different airway geometries so that their performance can be optimised.

 
What can we learn from a particle’s path through an inhaler?

The forces effecting the particles are difficult to control. We know that close-range adhesive forces by their very nature are extremely difficult to overcome using airflow or impact alone. Rather annoyingly, there is plenty of energy available – it’s just a bit tricky to transfer it to the particles. We need to explore all the possibilities that these particles present. By increasing the chance of impact we increase the chance of detachment. Unfortunately deagglomeration is too often the path less travelled by particles.

Like it