We use cookies to enhance your experience. By continuing to browse this website, you agree to our use of cookies. More information.
An emulsion is a two-phase system formed by droplets dispersed in another immiscible liquid. Many emulsions consist of a hydrophobic "oil" phase dispersed in a continuous water phase.
In addition, most emulsions contain at least one other substance, namely a surfactant, to stabilize the droplets to prevent aggregation and coalescence, and to help control whether the hydrophobic liquid is dispersed in the water phase, and vice versa.
The most common function of emulsions is to transport water-insoluble substances in a stable, finely dispersed form1, which provides many opportunities for designing advanced formulations in pharmaceuticals, cosmetics, and food/nutrition products.
The droplet size and size distribution (PSD) of the nanoemulsion have a significant impact on its rheological behavior, stability, functionality and safety2. In order to optimize these characteristics and control the size of various recipes and manufacturing settings, different manufacturing methods have been designed.
The high-pressure homogenization (HPH) that we focus on here dates back to the early 1900s when Auguste Gaulin invented milk homogenization. It is currently one of several so-called "high-energy" methods; others include microfluidization and ultrasound. Treatment 3, 4 Jet dispersion and high-amplitude ultrasonic treatment 5.
In a typical HPH process, first the rotor/stator homogenization is used to prepare the coarse emulsion, which is then used as the HPH feed.
In HPH, it then passes through a small nozzle, as shown in Figure 1, which causes high shear, droplet elongation and turbulent fluctuations under high pressure (up to several kilobars), which together cause the droplets to break into smaller droplets.
The pressure level that determines the energy input is the main process parameter to adjust the droplet size characteristics. Although significant progress has been made in understanding the mechanism of HPH particle size reduction6, there are still important challenges in concentrated emulsions7 and complex formulations found in practice7.
Aspects such as "overprocessing", the competition between droplet fragmentation and coalescence, and their dependence on surfactant properties and concentration are still poorly understood.
Figure 1. Schematic diagram of the high-pressure homogenization process. Image source: https://commons.wikimedia.org/wiki/File:Homogenizing_valve.svg
In fact, in many HPH processes, multiple passes through the emulsion are required to obtain monodisperse droplets (narrow size distribution), and the process is usually performed in a circulating loop.
Therefore, for development purposes or to ensure correct process parameters during scale-up and manufacturing, the droplet size needs to be carefully characterized at different points in time.
However, the traditional sizing method requires sampling and dilution steps, which brings the risk of measuring unrepresentative samples, prevents possible process control and requires labor hours.
The representativeness of the sample is particularly important for the homogenization of unstable (intermediate) emulsions, where offline characterization can lead to unreliable results due to droplet growth, and in the process, the emulsion may be kinetically stable.
This will be explained later in this article. Therefore, online and real-time sizing is the most ideal method for monitoring the emulsification process.
The FDA defines process analysis technology ("Quality by Design", part of QbD) as a system that designs, analyzes, and controls manufacturing by measuring key quality and performance attributes of materials and processes in a timely manner (that is, during processing).
Therefore, PAT aims to improve process understanding and ensure constant final product quality through online/online measurement of product characteristics, and ideally provide feedback for process control and product quality assurance ("quality attribute-based design space").
In addition to better product quality and consistency, PAT can also increase the one-time pass rate, reduce waste, minimize batches of non-conforming products and shorten the production cycle time. PAT is also critical for continuous manufacturing processes, where real-time process control is critical, and enabling "real-time release testing" to directly and safely release products to patients.
Although QbD began to be used in nanomedicine8, there has been a lack of direct online measurement or control of size characteristics, even though size characteristics are a major quality attribute9.
The dimensional characterization of the nanoemulsion process is currently performed almost completely offline, for example using laser diffraction, analytical centrifugation or traditional dynamic light scattering (DLS).
For PAT, the disadvantage of these technologies is their offline nature: they require varying degrees of sample preparation, cannot (or are difficult) to be performed under flow conditions, the speed of real-time measurement is limited, and the measurement geometry does not allow easy process integration.
Standard DLS is one of the most popular offline tools. It measures the fluctuations of the scattered laser light caused by the Brownian diffusion of droplets; the frequency of these fluctuations reflects the particle diffusion coefficient D, which provides the size d (and size distribution) through the Stokes-Einstein relationship, d = KBT /(3πηsD), where ηs is the solvent viscosity.
Real online measurements cannot be performed using various current DLS solutions: they require (almost) static conditions to ensure that the intensity fluctuations of the measurement are only caused by diffusion and are not affected by flow, and they are limited to low turbidity samples to avoid multiple times Scattering, and relatively high turbidity is often encountered in industrial processes such as homogenization.
In order to overcome these limitations, InProcess-LSP developed NanoFlowSizer10, which uses an innovative "low-coherence interferometry" that uses NIR (1300 nm) broadband light to achieve spatially resolved dynamic light scattering (SR-DLS).
SR-DLS not only averages the scattered signal, but also instantly analyzes the scattered light and its fluctuations at different depths in the sample (Figure 2). This data contains information about steady motion due to flow and Brownian motion.
The latter contribution can be extracted from SR-DLS measurements under suitable flow conditions and used to obtain dimensional features (Figure 3).
In addition, automatic depth analysis that excludes multiple scattered light can measure samples with turbidity levels that far exceed the current state-of-the-art DLS level (see Figure 4).
Fast measurement (~10 seconds) and online modules for flow rates from ~mL/min to more than 300 L/hr and offline measurement allow multiple applications of the instrument.
Figure 2. Traditional DLS and spatially resolved DLS. Left: Standard DLS only detects average scattered signals. Right: SR-DLS detects scattered light fluctuations as a function of depth. Image source: InProcess-LSP
Process monitoring using this instrument is achieved by coupling the probe unit with a special adapter, which includes a suitable flow cell (from millimeters to 2 inches inner diameter) into the process flow for direct measurement of the emulsion processing that is flowing .
Therefore, during the flow, real-time size characterization is achieved with the full concentration used in the process. This provides "immediate" and continuous information about the change in droplet size characteristics during the emulsification process and under different conditions.
The instrument can be applied from small-scale laboratory/pilot-scale processes to large-scale production pipelines.
Figure 3. In-depth analysis of measured intensity fluctuations and NanoFlowSizer's patented analysis method provides turbid and flowing suspension particle size characteristics. The upper middle part shows the single and multiple scattering states; the latter uses the NFS algorithm for effective filtering. The lower middle part shows the influence of flow on the correlation function, from which the Brownian motion component and flow can be analyzed. Image source: InProcess-LSP
For polystyrene particles, the turbidity range that the NFS instrument can operate is shown in Figure 4. The data points show the measured value of the NFS size at different concentrations, matching the size specified by the supplier (horizontal dashed line) for all concentrations used.
The line indicating the maximum size that can be reached at each concentration represents a constant turbidity level, and NFS can obtain a sufficient single scattering signal maximum. It is calculated using the Mie theory and the "hard ball" structure effect, and is suitable for high-concentration polystyrene11.
The dotted line represents the typical turbidity levels of other DLS technologies and illustrates the extended size/concentration range of NFS. Please note that the shape/range of the accessible range depends on the dispersive optical characteristics, see Figure 8b.
Figure 4. The size-concentration scheme of the polystyrene reference standard using SR-DLS. Use the Mie-Percus-Yevick model 11 to calculate the line, which represents the maximum turbidity achievable by NFS and other technologies. Image source: InProcess-LSP
Although Figure 8 shows that NFS can be used to measure relatively large turbidity samples (large size and/or concentration), it is well known that particle dispersion measured by NFS can modify high volume fraction interactions due to "crowding" or other reasons.
The measured diffusion coefficient can be used "as is" for qualitative process monitoring, but for direct quantitative calibration of the "blocking factor", d/deff (φ) can be established using the dilution of the collected sample. Here deff (φ) is the "effective" size of the volume fraction φ, and d is the actual size measured under dilution (φ <∼0.01).
Note that the common way to explain "impeded diffusion" is to use the generalized Stokes-Einstein relationship d = KBT/(3πηD), where emulsion viscosity is used instead of solvent viscosity. As we all know, this method may have fundamental flaws and may only be suitable for specific situations12.
Two examples of online monitoring of two different emulsion systems are described. The first involves model oil/water (O/W) emulsions prepared at three different oil volume fractions, with a maximum of φ = 0.22, each of which is homogenized in a separate run.
The second example uses an unstable drug emulsion system consisting of an organic solvent phase dispersed in an aqueous phase (OS/W).
Figure 5 shows the principle diagram of the homogenization circuit in the two cases of integrated NFS. Both contain a circulation loop in which the emulsion is continuously homogenized using a laboratory-scale homogenizer.
In both cases, the temperature rise of the emulsion in the homogenizer is relieved by a heat exchanger placed after the HPH. During this process, a sensor integrated near the flow cell is used to monitor the temperature and use it as an input to the NFS to calculate the appropriate solvent viscosity.
During different homogenization processes, NFS analyzes the droplet size in real time and continuously flows in the cell.
Figure 5. Schematic HPH flow chart of integrating NFS as a PAT tool to determine droplet size (a) Configuration of three O/W emulsion runs (b) Configuration of unstable OS/W emulsion homogenization. Image source: InProcess-LSP
An important aspect of using NFS as a PAT tool is positioning and its integration in the process. First of all, it is necessary to obtain process/product-related droplet size information, which will affect the choice of position relative to HPH.
Secondly, in order to obtain the best online measurement performance, a pulsation-free laminar flow, fully developed flow is required. Since the flow directly at the HPH outlet is pulsating, two different solutions are used. For O/W emulsion operation, NFS is located at the inlet of HPH. The advantage is that the characteristics of the emulsion in the reservoir can be measured.
For unstable OS/W homogenization, place the instrument in the heat exchanger, fully reduce the pulsation, stop the flow, and study the emulsion stability.
The three emulsions used in continuous operation consist of purified sunflower oil (Sigma-Aldrich) with volume fractions φ = 0.055, φ = 0.11 and φ = 0.22 (5, 10 and 20 wt%) in 1 wt% tap water Tween 20. First homogenize each coarse emulsion at 200 bar until a stable size and polydispersity is reached, then increase the pressure in steps of 200 bar, each time the dimensional characteristics become constant, up to 800 bar.
The actual setting corresponding to Figure 5a is shown in Figure 6. The NFS is connected to a 3/4-inch inner diameter flow cell with a sight glass, and the flow rate through HPH (Panda 2000, GEA) is about 8 L/h. The emulsion flow through the NFS is provided by a stable non-pulsating micro gear pump.
Figure 6. Using NFS to set up online monitoring of droplet size during O/W emulsion homogenization. Image source: InProcess-LSP
In order to evaluate the possible influence on the diffusion rate due to the high volume fraction, samples were taken from the reservoir at each setting of the quiescent state. The dilution series of these samples were prepared using 1% Tween solution, with a concentration φ <0.005, and measured by NFS in a vial (no flow).
In addition, the dilution series of another nanoemulsion (commercially available Intralipid, Z-av size 340 nm) with the same volume fraction (φ = 0.22) as the most concentrated sunflower O/W emulsion was examined to evaluate the two differences But the "blocking factor" d/deff (φ) between similar nanoemulsions.
For proper data analysis, the influence of surfactants on solvent viscosity is considered. Following the "Jones-Doll" formula 13 ηs = ηs (c), a limited concentration of Tween 20 enhances this viscosity. For our concentration (c0 = 8.2mM >> critical micelle concentration), this will get ηs (c0) = ηs,0 ≃ 1.1ηwater but a small part of Tw20 is absorbed on the droplet interface, as the homogenization process The droplet size decreases and increases.
The net "free" Tw20 concentration in the solvent is calculated from the specific droplet surface area 1 S ≃ 9φ/d and the Tw20 specific molar surface area AT,M = 3.6 • 105 14, giving c(φ, d) = c0 − (9φ/AT, Md). This gives the droplet size-dependent solvent viscosity ηs (d) and the implicit size dependence of the Stokes-Einstein form. In order to convert the diffusion coefficient D into the droplet size in real time, we used the approximate value ηs (φ, d) ≃ ηs,0 − ηwater [δ(φ)/d], where δ(φ) is the offset related to the volume fraction quantity. The resulting conversion d ≃ [δ(φ)ηwater /ηs,0] [kBT/3πηs,0D] can be integrated into the real-time analysis of the NFS system2.
1 Formally S = 6φ/d3,2, where d3,2 is the Sauter average diameter. For the measured polydispersity level, the ratio of d3,2 to the Z-av diameter d is ~2/3, giving the stated form.
2 This includes the temperature dependence of δ(φ), ηs, 0 and ηwater.
The drug emulsion consists of organic solvent droplets (~3 vol%, dissolved API) dispersed in an aqueous solvent (the effect of additives on viscosity is estimated to be negligible). The turbidity is very similar to the above 5 vol% O/W emulsion.
Use the settings in Figure 5b to homogenize and monitor the emulsion using Panda1000 with a flow rate of about 20 L/hr. Here NFS is combined with a 1-inch flow cell. During this process, the pressure in HPH remains constant. After 39 minutes, stop the flow for about 4 minutes to check the stability of the emulsion when there is no flow.
Figure 7a shows the monitoring results of the 5 vol% O/W emulsion during the gradual pressure change. After the pressure was increased to 400 bar, a rapid decrease in the droplet size (Zav) was observed, changing from about 450 nm to slightly below 350 nm in 50 minutes, as shown in Figure 7a.
After the pressure was increased to 600 bar, the droplet size was reduced to a minimum size of about 280 nanometers. After about 30 minutes, the pressure was increased to 800 bar, resulting in a constant droplet size of 250 nm. The reduction in PDI from the beginning to the end of the process can also be observed in Figure 7a. In addition, Figure 7b shows the PSD at different time points.
Compared with the close to the end time point, a significantly wider particle size distribution can be observed at the early time point. Over time, the PSD becomes narrower and the average peak shifts to a smaller droplet size. These results confirm the expectations of droplet size and polydispersity changes during homogenization. The offline sample taken from the reservoir was measured in a glass vial with no flow for comparison, as shown by the circle in Figure 7a, and confirmed the real-time online data.
Figure 7. (a) Online monitoring of the size characteristics of 5 vol% O/W emulsion after homogenization under different pressures (b) Droplet size distribution at different time points. Different line colors represent different time points (blue Color: from ± 2 minutes to ± 60 minutes, green-red: from ± 70 minutes to ± 170 minutes). Image source: InProcess-LSP
Figure 8a shows the terminal "steady-state" droplet size data at different concentrations obtained from all HPH runs, illustrating the "impeded diffusion" of the 11 vol% and 22 vol% data. The size and polydispersity of all volume fractions have a similar tendency to decrease, but the droplet size of large concentration is slightly larger, and the polydispersity is slightly smaller.
These data prove that SR-DLS can measure samples in a wide concentration range and in the presence of flow. The size-concentration diagram of the nanoemulsion in Figure 8b (calculated using the optical parameters of the emulsion, lipid RI=1.47) illustrates the change in droplet size during the three runs.
All three nanoemulsions are in the turbidity range that other DLS methods cannot achieve.
Figure 8. (a) The relationship between terminal droplet size and polydispersity and homogenization pressure, obtained by online monitoring of three different volume fractions of HPH operation (b) Compared with other DLS, using SR-DLS to determine The size-concentration scheme diagram system of lipid/water emulsion size. The black dots correspond to the initial (top) and final (bottom) droplet sizes during the HPH run of different volume fractions. Image source: InProcess-LSP
The results in Figure 9 show that for the volume fraction φ = 0.055, the size correction due to hindered diffusion is almost negligible, but for φ = 0.109 and φ = 0.22, a significant reduction in diffusion is observed.
Therefore, in order to obtain quantitative particle size information for these emulsions, a hindered diffusion correction factor must be applied. For this type of emulsion, it has significant polydispersity and interaction dominated by "excluded volume". The size-independent calibration factor can sufficiently (within about 5%) to convert the diffusion coefficient measured in the concentrated sample online Is the actual droplet size.
The obtained volume fraction is very similar to the correction factor of the lipid sample with the same volume fraction of the most concentrated sunflower emulsion, confirming the consistency of the correction factor of similar formulations.
For comparison, the correction factor estimated from the generalized Stokes-Einstein equation, which is based on the ratio of solvent viscosity and emulsion viscosity at different volume fractions, is also shown. The difference with the actual observed scale factor indicates that this viscosity-based correction is difficult to explain the hindered diffusion universally.
Figure 9. The "impeded diffusion" correction factor for the quantitative size of concentrated emulsions determined by comparing diluted and undiluted samples. Image source: InProcess-LSP
Figure 10 shows the monitoring of the Z average of the nanodroplets during the organic solvent/water emulsion emulsification process in about 40 minutes. At the beginning of the process, the average droplet size was about 200 nm. The droplet size then showed a slight decrease during the long-term homogenization, most notably after about 20 minutes. After stopping the flow for 39 minutes, the droplet size showed a significant increase from about 170 nm to about 330 nm. When the flow starts again, the droplet size decreases almost immediately and begins to backtrack the slow decreasing trend observed before turning.
Figure 10. Online measurement of Z-average droplet size using NFS during homogenization and flow interruption of OS/W emulsion. Image source: InProcess-LSP
The size increase (and decrease) seen before (and after) the rerouting is not a measurement artifact: the NFS algorithm provides a separate analysis of flow and diffusion (Brownian) motion for each measurement, while the latter provides dimensional characteristics.
This data clearly shows how to use the NFS online monitoring function to better understand the homogenization of emulsions that exhibit complex and unstable behavior.
Due to the unstable nature of emulsions, it is impossible to reliably measure the droplet size during processing offline in practice. Therefore, in this case, the use of PAT tools for dimensional characterization is particularly valuable.
Introduced a new type of instrument (NanoFlowSizer) and method for online nanoparticle classification to achieve rapid and effective process control and formulation development.
The results of rapid monitoring of the nanoemulsion size evolution during the two high-pressure homogenization processes show that (i) the size of nano-droplets can be measured in real time online without dilution and (ii) the instrument allows size characterization in the case of failure of the standard method. For example, due to limited stability.
Spatally Resolved DLS provides a wide range of accessible concentration and turbidity requirements and therefore provides a good understanding of the possible impact on the diffusion rate in undiluted samples.
This effect, as well as the dynamic viscosity change during processing, can be calculated or empirically determined and effectively combined to obtain real-time size monitoring.
The ability to continuously obtain droplet/particle size information during processing to understand and control process and formulation parameters highlights the added value of the instrument from R&D to commercial manufacturing.
This information is derived from, reviewed and adapted from materials provided by InProcess-LSP.
For more information on this source, please visit InProcess-LSP.
This article was written by Rut Besseling of InProcessLSP.
Please use one of the following formats to cite this article in your paper, essay, or report:
InProcess-LSP. (2021, September 30). Real-time droplet size monitoring of nanoemulsion during high-pressure homogenization. AZoNano. Retrieved from https://www.azonano.com/article.aspx?ArticleID=5679 on November 25, 2021.
InProcess-LSP. "Real-time droplet size monitoring of nanoemulsion during high-pressure homogenization". AZoNano. November 25, 2021. <https://www.azonano.com/article.aspx?ArticleID=5679>.
InProcess-LSP. "Real-time droplet size monitoring of nanoemulsion during high-pressure homogenization". AZoNano. https://www.azonano.com/article.aspx?ArticleID=5679. (Accessed November 25, 2021).
InProcess-LSP. 2021. Real-time droplet size monitoring of nanoemulsion during high-pressure homogenization. AZoNano, viewed on November 25, 2021, https://www.azonano.com/article.aspx?ArticleID=5679.
Do you have any questions about this article?
AZoNano talked to Dr. Jinian Yang about his participation in research on the benefits of flower-like nanoparticles on the performance of epoxy resins.
We discussed with Dr. John Miao that this research has changed our understanding of amorphous materials and what it means for the physical world around us.
We discussed NANO-LLPO with Dr. Dominik Rejman, a wound dressing based on nanomaterials that promotes healing and prevents infection.
The P-17 stylus profiler surface measurement system provides excellent measurement repeatability for consistent measurement of 2D and 3D topography.
The Profilm3D series provides affordable optical surface profilers that can generate high-quality surface profiles and true color images with unlimited depth of field.
Raith's EBPG Plus is the ultimate product of high-resolution electron beam lithography. EBPG Plus is fast, reliable, and high-throughput, making it ideal for all your lithography needs.
AZoNano.com-AZoNetwork website
Owned and operated by AZoNetwork, © 2000-2021