The first direct observation of pharmaceutical non-classical crystallization

The first direct observation of pharmaceutical non-classical crystallization

Using the Ocean system, scientists achieve supersaturation in LPEM experiments, revolutionizing pharmaceutical crystallization


Original article by Jennifer Cookman, Victoria Hamilton, Simon Hall and Ursel Bangert

LPEM video showing the pre-crystallization process of flufenamic acid

Whereas classical crystallization deals with layer-by-layer growth of crystals, non-classical crystallization (NCC) involves multiple different crystallization pathways towards the formation of final stable crystals. Although NCC has been widely documented in research, there is still much to be explored regarding the intermediate stages of crystallization and their direct observation. This is especially true for small organic molecules like flufenamic acid (FFA), an anti-inflammatory drug used for the treatment of rheumatic disorders.

Using the DENSsolutions Ocean LPEM system, Dr. Jennifer Cookman from the Bernal Institute in the University of Limerick and her colleagues were able to capture the intermediate pre-crystalline stages of this small organic molecule. This research marks the first ever direct observation of a pharmaceutical material undergoing NCC, highlighting the rising value and importance of in-situ TEM techniques in the pharmaceutical industry. 

The observed processes of NCC

Crystallization is a fundamental process that occurs in nature to produce some of the most common materials in daily life, such as the popular active pharmaceutical ingredient (API) ibuprofen or FFA. Properties such as solubility and bioavailability are linked to the crystal structure of the active compound. Considering APIs are commonly polymorphic, it is important to understand the intermediate stages of their crystallization. Specifically, if we can identify polymorphs with more desirable properties in the intermediate stages of crystallization, then this opens the door to harnessing and potentially directing their formation.

In this study, Dr. Cookman and her colleagues observed in situ the processes involved in the nanoscale crystallization of FFA. As illustrated in the figure below, this process involves four stages: aggregation, coalescence into a metastable entity, nucleus formation, and finally, crystallization.

A summary of the observed processes involved in the nanoscale crystallization of FFA

The researchers observed that FFA begins as a collection of small independent pre-nucleation clusters (PNCs). These PNCs are essentially stable particle clusters that form prior to the nucleation of a solid phase. They were able to follow three notable aggregates of PNCs that each followed the same transformational events. Particularly, after aggregation, these PNCs each independently coalesced, or merged, and formed a metastable phase. After this, the densification and development of a nucleus occurs, leading to the formation of FFA crystals. The processes of coalescence and densification will be further discussed and depicted below.


The aggregation of the PNCs were shown to have occurred prior to the researchers’ initial observations. Therefore the primary transformation observed for the three aggregates was actually that of coalescence. In the image below, you can see clearly that for each of the three selected aggregates, the individual clusters merge to form one cohesive entity after approximately 3 minutes.

A time-lapse of each of the three aggregates of PNCs undergoing coalescence

Densification towards crystallization

Following coalescence is the densification and development of a nucleus. This nucleus is formed by the successive sacrifice of surrounding material, leading to the formation of a new crystalline-like object, significantly more electron dense than before. Whereas coalescence took around 3 minutes, this densification occurred rapidly in under 10 seconds. The image and three videos below depict this rapid pre-crystallization process of FFA. 

A frame-by-frame summary of the three aggregates illustrating the pre-crystallization process of FFA

Aggregate 1

Aggregate 2

Aggregate 3

Novelty in findings

This research contributes academically in that the direct observations reported for the crystallization of FFA reveal insightful new information about the potential pathways towards crystallization. Moreover, it highlights the need to further investigate the nucleation and resulting crystallization of other small organic molecules via in situ techniques such as LPEM. LPEM presents itself as a required and complementary tool to not only comprehend but also probe chemistry at the nanoscale. This is true especially in regards to the crystallization of pharmaceutical ingredients, in which the development of the end product highly depends on controlling at the molecular building block level. 

The novelty of this research also lies in that it sheds light on the crystallization and nucleation of pharmaceutical products, providing the necessary information to further refine industrial-scale processes. If we can observe and understand the crystallization pathways that small organic molecular crystals like FFA take, we can better streamline production activities and develop effective manufacturing processes for generic drugs. It is precisely our goal at DENSsolutions to enable researchers like Dr. Jennifer Cookman to continue to bridge gaps in research using our solutions and uncover results that can impact this world, in the pharmaceutical industry and beyond.  

DENSsolutions Jennifer Cookman

“The DENSsolutions Ocean holder is a simple solution to native environment metrology that has the potential to revolutionize how we view pharmaceutical crystallization.”


“The DENSsolutions Ocean holder is a simple solution to native environment metrology that has the potential to revolutionize how we view pharmaceutical crystallization.”

Dr. Jennifer Cookman
Post Doctoral Researcher | University of Limerick

Original article:


Discover our latest LPEM solution, Stream:

Discover more publications made possible by Ocean:

Do you want to receive great articles like this in your mailbox? Subscribe to our newsletter.

Achieving mass transport control with the award-winning Stream system

Achieving mass transport control with the award-winning Stream system

The on-chip flow channel of the Stream system allows for full control over pressure, flow rate, liquid thickness and electric potential

Original article by Anne France Beker, Hongyu Sun, Mathilde Lemang, Tijn van Omme, Ronald G. Spruit, Marien Bremmer, Shibabrata Basak and  H. Hugo Pérez Garza

The liquid phase transmission electron microscopy (LPTEM) community faces numerous challenges when performing in situ electrochemical studies inside the TEM. From a lack of control over the flow and liquid thickness, to limited experimental flexibility and reproducibility, these challenges have posed considerable limitations on research. As a result, DENSsolutions has developed an in situ LPTEM solution that addresses each and every one of these challenges – the Stream system. Due to its unique on-chip flow channel design, users can effectively control experimental conditions such as pressure, flow rate, liquid thickness, electrical potential and bubbles. STEM videos are shown below to demonstrate these advantages and visualize the in situ growth of copper with multiple morphologies.

Because you can independently control the pressure at the inlet and outlet of the Stream Nano-Cell, you can control the absolute pressure in the microfluidic channel. This state-of-the-art design consequently gives you full control over the flow and the bulging of the windows, and therefore the liquid thickness. As a result, spatial resolution is improved, enabling meaningful electron diffraction and elemental mapping in liquid. You can accurately define the mass transport and control the electric potential, granting you complete access to the full kinetics of the reaction.

The in situ LPEM study

In order to exhibit the benefits of the system, copper dendrites were grown and characterized in situ. After the electrodeposition of the copper, EELS and EDS characterization were performed with copper inside the viewing area. Furthermore, high resolution images and diffraction patterns of the grown copper dendrites were recorded using the TEM.

Removal of beam-induced species

A major issue when performing LPTEM experiments with an electrolyte is the undesired influence of the electron beam. In this experiment, the electron beam interacts with the copper electrolyte. However, because you can control the flow of the liquid, you can remove or flush away any unwanted beam-induced species from the region of interest (i.e. window, sample or electrodes). This is displayed in the STEM recording below with the flow moving from right to left.

STEM movie showing debris being flushed

Bubble dissolution

It is important in LPTEM to assure that the cell stays wet. However, when bubbles form, the cell starts to dry out. The Stream system was developed with this in mind, offering a solution to this challenge. Specifically, because you can control the absolute pressure in the microfluidic channel, you can remove unwanted gas bubbles by setting the pressure high. At higher pressures, the size of the bubble decreases until it disappears and vice versa. The dissolution of a bubble that was formed during this copper experiment is shown in the STEM video below.

STEM movie showing bubble dissolution

In situ growth of copper dendrites

The growth and stripping of copper was completed a few times via cyclic voltammetry. The cycles begin with copper reduction, corresponding to the growth of the copper dendrites. Next, oxidation takes place, corresponding to the copper dendrites being stripped. Interestingly, you can see in the STEM video below that after reduction, the dendrites are thicker whereas after oxidation, the dendrites become much thinner.

STEM movie showing 5 cycles of copper growth and etching

Liquid thickness control 

In order to perform high resolution imaging, it is important in LPTEM that the liquid thickness is kept low. Aside from high resolution imaging, controlling the liquid thickness is extremely important when performing analytical techniques like EDS, EELS and electron diffraction. Ideally, the liquid should be limited below the beam broadening, which is normally expected to happen around 500nm of liquid thickness. With this in mind, we designed our Nano-Cell such that the thickness stays below the beam broadening threshold based on the spacer thickness and the maximum bulging of the windows. In the figures below, the elemental mapping and electron diffraction of the electrodeposited copper are presented. 

Elemental mapping - Anne article

EDX elemental mapping showing the spatial distribution of b) the copper dendrites and c) the platinum electrode 

Electron diffraction Annette article

TEM image of the copper dendrites on the electrode in e) and the corresponding SAED patterns in liquid phase in f)

Complete flow control

Controlling the flow also has other important advantages that are expanding possibilities in research. Namely, the ability to manipulate the flow rate allows you to control the morphology. You can see in the STEM image below that when flow is applied, the copper grows in a continuous layer with more copper being deposited. On the other hand, without flow, the copper nuclei grow isolated. This is direct proof that the unique flow-control feature of the system allows you to control the kinetics of an electrochemical reaction.

Morphology of copper with and without flow using the Stream system

Conclusively, this research highlights the unique capabilities of the award-winning Stream system, proving its potential to enable and boost research in various application fields, ranging from battery research and fuel-cells to corrosion and electrocatalysis.

Original article:

More about Stream:

Do you want to receive great articles like this in your mailbox? Subscribe to our newsletter.