Method and apparatus for in-process particle size measurement of a nano-suspension under flow

Abstract

To provide a novel method for determining a property of nanoparticles in a flowing nanosuspension.SOLUTION: A method for monitoring a property of nanoparticles in a flowing suspension includes a step of preparing a sample comprising a flowing suspension. The method further includes a step of non-invasively monitoring a particle size distribution of nanoparticles of the flowing suspension using Fourier domain low-coherence interferometry and FDLCI, in which the monitoring comprises deriving a time and optical path length resolved LCI light scattering signal I(t, z) from time resolved LCI wavelength spectra of interference and deriving information indicative of the size of the particles in the sample based on the time and optical path length resolved LCI light scattering signal using optical path length resolved temporal autocorrelation functions or optical path length resolved frequency power spectra of a spatiotemporal FDLCI signal.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to the measurement of the particle size of a colloidal suspension. More specifically, the present invention relates to a method and apparatus for non-invasive real-time in-process measurement of the particle size distribution, flow, and physical properties of a flowing colloidal suspension by optical path length resolved photon correlation spectroscopy.

Background Art

[0002] Due to the unique advantages offered by nanoparticle (NP) products and formulations, the synthesis and processing of nanoparticles / colloids (in the particle size range of approximately 1 to 1000 nm) have become widespread in various industries over the past few decades. For example, in the pharmaceutical industry, therapies formulated as NPs can offer better pharmacokinetic properties, sustained release, and targetability. In food and cosmetics, colloids that occur naturally or are present due to formulations are also abundant. The increasing efforts in NP product development and manufacturing, as well as the growing demand for monitoring NP-involved processes, such as to characterize the effects of process variations during development or to ensure quality during normal production, have led to an increasing need for in-line non-invasive methods to characterize NPs, particularly NPs in suspension, during these processes. In many cases, the particle size and particle size distribution (PSD) of NPs, if not decisive, are major quality attributes. Therefore, there is a strong need for a method for in-line particle size characterization of NP suspensions during processing.

[0003] Current common methods for monitoring NP particle size or PSD in a process involve taking a manual sample of the suspension followed by off-line analysis. This has significant disadvantages such as excessive feedback time for process adjustment, risk of product quality degradation and batch rejection, uncertainty in the representativeness of the measurement due to and / or resulting from reduced suspension stability after sampling and / or sample preparation. In unstable suspensions, off-line analysis may not be fully possible. Sampling also carries the risk of contamination or product sterility, product loss and high cost, further increasing the need for non-invasive in-line NP particle size characterization.

[0004] There are a wide variety of methods currently available for offline (sample) analysis of the particle size and PSD of NP suspensions, but there are limited or no opportunities to use these methods non-invasively during the synthesis or processing of the suspension. Examples include analytical centrifugation / photosedimentation, single particle mass analysis, or other separation-based methods such as size exclusion chromatography, which are inherently invasive. The most frequently used offline NP measurement technique uses the optical detection of the Brownian motion of the suspended colloids, from which the particle size characteristics can be obtained by the Stokes-Einstein relationship. The best-known of these techniques is Photon Correlation Spectroscopy (PCS) (Berne, B & R Pecora, 2000. Dynamic Light Scattering: With Applications to Chemistry, Biology, and Physics, Dover Publications), which covers various methods by which the Brownian motion, and thus the particle size, of a population of suspended NPs is characterized by the measurement of the temporal fluctuations and correlations of the light scattered from the suspension. Other such methods include nanoparticle tracking analysis (NTA) (see, for example, U.S. Patent No. 7,751,053, U.S. Patent No. 9,341,559, etc.), which uses digital video microscopy techniques to track the Brownian motion of many individual suspended NPs, and more recently developed, Differential Dynamic Microscopy (R Cerbino, 2008. Phys Rev Lett 100:188102), a population scattering method based on video microscopy that probes the collective diffusion of the colloids.

[0005] There are other optical methods that detect the average scattered signal (such as turbidity) or the average angle-resolved scattered signal instead of Brownian dynamics, but these are usually sub-optimal for measuring the PSD in the colloidal particle size range (see, for example, U.S. Patent No. 5,377,005, U.S. Patent No. 6,831,741, and U.S. Patent No. 5,438,408, etc.). For example, these methods are indirect, require calibration by a reference method, and mainly provide average particle size information. In this context, it should be emphasized that in practice, the preparation of a suitable calibration material (a suspension containing NPs of different known particle sizes) can be very complex or even impossible. Another method is Photon Density Modulation (U.S. Patent No. 5,818,583 and U.S. Patent No. 6,480,276), and similar drawbacks apply to this as well.

[0006] The above-described methods based on the characterization of Brownian motion are essentially limited to offline analysis for the following reasons. (i) These methods require a suspension with a low turbidity level (scattering coefficient μ s << 1 mm -1 ) that is different from the turbidity levels typically occurring in industrial processes, (ii) these methods require no flow in the suspension, which is a major obstacle to in-line analysis, and (iii) the analysis time of these methods is at best in minutes and is too slow for process monitoring and control.

[0007] Several methods have been devised to overcome the limitations regarding the turbidity of suspensions. One example is cross - correlation PCS, which suppresses multiple - scattered light resulting from the measurement of highly turbid media, but this method is not suitable for in - line measurements. Another example is Diffusing Wave Spectroscopy (DWS), for example, see U.S. Patent No. 6,831,741 and Pine, DJ et al., 1990 Diffusing - wave spectroscopy: dynamic light scattering in the multiple scattering limit Journal de Physique, 51(18), pp. 2101 - 2127, which utilizes multiple scattering on the one hand. DWS can be configured for in - process measurements, but provides only average particle size information and cannot handle non - stationary (flowing) suspensions under processing conditions.

[0008] Other inventions are based on different extensions of PCS for turbid media. U.S. Patent No. 5,094,532 discloses performing PCS using a 'heterodyne' signal formed by mixing light back - scattered from a sample with light reflected from the tip of a measurement probe. This 'heterodyne back - scattered PCS' enables the characterization of higher - concentration suspensions compared to standard PCS, but is also not suitable for non - stationary (flowing) suspensions under processing conditions. European Patent No. 2270449 discloses the use of Low Coherence Interferometry (LCI) in PCS. LCI - PCS also measures a heterodyne interference signal, using low - coherence light with a coherence length ξ as illumination, and the back - scattered light is mixed with reference light that has traveled a controlled optical path length in the reference optical path. LCI provides a 'coherence gate' to limit the detection of light scattered from the sample to light with an optical path length equal to the controlled path length ±~ξ. LCI - PCS fluctuation and correlation analysis at short, fixed optical path lengths where single scattering applies further improves PSD measurements in concentrated samples, but is also not suitable for in - line application to flowing NP suspensions.

[0009] U.S. Patent No. 6,738,144 discloses the measurement of photon optical path length distribution in a regime of multiple scattering for particles in a non-colloidal particle size range with a diameter greater than about 2 - 5 μm using LCI.

[0010] Previously, there has been no method for the real-time direct measurement of the particle size distribution in a regime of submicron (colloidal) particle sizes in a dynamic process environment with flow on a process-related time scale.

Summary of the Invention

[0011] One object of the present invention is to achieve improved detection of colloidal particle characteristics in a suspension. This object is a step of non-invasively monitoring at least one of the particle size and particle size distribution of colloidal particles in a flowing suspension using Fourier domain low coherence interferometry, FDLCI, the monitoring step comprising the steps of deriving a time and optical path length resolved LCI light scattering signal I(t,z) from the time-resolved LCI wavelength spectrum of the interference, and the optical path length, z, decomposition time autocorrelation function G(τ,z) of the time and optical path length resolved LCI light scattering signal I(t,z) as well as the optical path length resolved frequency power spectrum of the time and optical path length resolved LCI light scattering signal I(t,z)

Number

[0012] The time and optical path length resolved signal I(t,z) obtained from the FDLCI enables taking into account the difference in the time variation of the LCI signals resolved virtually simultaneously at different optical path lengths in the suspension. These different optical path lengths can on average correspond to different depths in the suspension from which the photons are scattered. Thus, the influence of the flow, which can be different at such different depths when calculating the particle size and PSD in the suspension, can be separated from and corrected for the time variation due to Brownian motion. Thus, the present method can provide non-invasive in-line particle size and PSD characterization of flowing colloidal suspensions in industrial applications, for example, during synthesis in a batch reaction vessel (where the suspension can be circulated during the process) or in a continuous process. Furthermore, the in-line particle size and PSD characterization can be performed independently of the concentration of the colloidal particles without calibration by a reference method. In addition, the high-speed acquisition enabled by FDLCI, together with the high-speed processing of the optical path length resolved data, also enables the characterization to be performed on a process-related time scale.

[0013] Information indicating the particle size distribution of the colloidal particles can be obtained from at least one optical path length (z) resolved temporal autocorrelation function G(τ,z), where τ represents the delay time of the time and optical path length resolved LCI scattering signal I(t,z), or equivalently, from at least one power spectrum [Number] which characterizes the time variation of I(t,z) caused simultaneously by the flow and Brownian motion of the suspended particles. One object of the present invention is to provide a method for separating and characterizing the signal variations associated with Brownian motion, thereby extracting the particle size characteristics of the flowing suspension.

[0014] Those skilled in the art will understand that the above features can be combined in any way considered useful. Further, the modifications and variations described with respect to the system are equally applicable to the method and the computer program product, and the modifications and variations described with respect to the method can equally be applied to the system and the computer program product.

[0015] Hereinafter, aspects of the present invention will be described by way of example with reference to the drawings. The drawings are schematic and may not be drawn to scale. Throughout the drawings, like items are denoted by the same reference numerals.

Brief Description of the Drawings

[0016]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5A

Figure 5B

Figure 6

Modes for Carrying Out the Invention

[0017] With reference to the accompanying drawings, specific exemplary embodiments will be described in more detail.

[0018] The problems disclosed in this specification, such as detailed structures and elements, are provided to assist in an overall understanding of the exemplary embodiments. Thus, it will be apparent that the exemplary embodiments can be practiced without those specifically defined problems. Also, well-known operations or structures are not detailed so as not to obscure the description with unnecessary details.

[0019] This specification discloses several examples and embodiments of methods and apparatuses for optical path length resolved low coherence photon correlation spectroscopy. These methods and apparatuses may enable non-invasively monitoring at least one of the particle size distribution, flow, and optical properties of a colloidal suspension, for example, simultaneously in real time during their synthesis and / or processing. Further, the methods and apparatuses can be designed and used without requiring calibration by a reference method.

[0020] In certain embodiments, the average particle size and particle size distribution can be obtained without calibration or knowledge of prior suspension characteristics (such as the concentration of colloidal particles). Additionally, since it is possible to measure at high concentrations and during flow, the method is suitable for industrial suspensions and in-line applications. The method can obtain an optical path length resolved scattering signal I(t,z) from scattered particles in a fluid that is virtually instantaneously resolved at a high sampling frequency for each time t using high-speed FDLCI. This includes a known technique for obtaining a spectrum I(t,λ) at time t and wavelength λ of the mixing of the reflected reference light and the light backscattered from the suspension from an FD low coherence interferometer, and several signal processing steps including an inverse Fourier transform that yields a complex optical path length resolved FDLCI signal I(t,z). The method is described herein using the complex optical path length resolved FDLCI signal I(t,z), but it is also possible to achieve the same result using the modulus. In such a case, the remainder of the method would be carried out similarly. According to the present disclosure, the method obtains from I(t,z) a correlation function G(τ,z) or a power spectrum

Number

[0021] FIG. 1 shows a schematic diagram of an FDLCI apparatus. The apparatus can comprise the following components. The light source 1 provides low-time-coherence (broadband) high-spatial-coherence light. The light source 1 c and bandwidth (usually considered the full width at half maximum FWHM of Δλ c ) can be a supercontinuum light source or a superluminescent diode characterized thereby. The low-time coherence of the light source is preferred to create a spatial 'coherence gate' for LCI to resolve the final backscattered light for a specific optical path length. The central wavelength λ c (which can be different from the maximum intensity wavelength of the asymmetric light source spectrum) can be in the range of 300 - 2000 nm, preferably 500 - 1500 nm. The bandwidth Δλ c can be several hundred nm, typically at least 50 nm, to achieve sufficient resolution for the final optical path length decomposition signal. In a specific implementation example, the central wavelength of the light source is λ c = 1300 nm and the bandwidth can be FWHM = 170 nm.

[0022] The light from the light source 1 can be coupled to the fiber 16a and guided through the direction element 6. This direction element 6 may include an optical circulator or an isolator, and serves to guide the return light to the spectrometer unit 2 and prevent it from re-entering the light source 1. The light from the light source then continues to be coupled to the fiber collimator or fiber focus 7 of the interferometer unit 17.

[0023] The spectrometer 2 receives and detects the mixed light of the backscattered light and the reflected light through the direction element 6. In a typical configuration, the light from the fiber 16b is collimated into a beam, and this beam is directed onto a diffraction grating or other dispersive element, and then guided using an appropriate optical system onto a linear camera / diode array detector (such as a charge-coupled device, CCD, or complementary metal-oxide semiconductor, CMOS) at an appropriate acquisition frequency. The maximum acquisition frequency (especially together with the optical system and the sample scattering characteristics) determines the minimum particle size and the maximum flow rate that can be reliably measured. The acquisition frequency can be, for example, about 50 kHz in a particular embodiment. Alternatively, for example, when using a swept wavelength light source instead of a broadband superluminescent diode (SLD) light source, a single balanced detector may be used instead of the dispersive element / array detector.

[0024] The data acquisition and processing unit 4 may include standard acquisition hardware (such as a frame grabber, data acquisition card, etc.) and at least one known data processing unit such as a CPU, GPU (graphics processing unit), or FPGA (field programmable gate array). The at least one processing unit is configured to receive the spectral interference data obtained from the spectrometer unit 2 and perform, in particular, primary and / or secondary processing on the data. The data processing may include primary processing involving standard processing steps including DC background removal and inverse Fourier transform, in which each raw LCI interference spectrum (i.e., the LCI signal intensity for each wavelength at time t) is converted into the corresponding optical path length resolved complex LCI signal I(t,z).

[0025] The secondary data processing step is to evaluate the optical path length (z)-resolved autocorrelation function G(τ,z) of I(t,z) or equivalently the power spectrum

Number

Number

Number

Number

Number

Number

[0026] The control unit 5 can convert the output of data processing (e.g., scattering intensity, particle size, and polydispersity index, PDI) into two sets of electrical signals. One set can control data acquisition settings (such as frequency) and measurement settings of the interferometer (such as the intensity of the reference arm light). The other set can represent the difference between the measured physical properties of the suspension or, for example, the operating parameters for synthesizing or processing the suspension so that these properties deviate from the target, in particular PS, PDI, and PSD, to control and / or adjust the measured properties and the target properties. The operating parameters can include any one or more of temperature, reactant concentration, stirring speed in a batch process, or various upstream / downstream process parameters in a continuous flow process.

[0027] The interferometer unit 17 can be implemented in the measurement probe 18. The interferometer unit 17 can include a fiber collimator or a focusers 7. Additionally, the interferometer unit 17 can also include various other components such as a beam splitter 8, an attenuator 9, a dispersion compensator or a focusing lens 10, a reflector 11, a controllable shutter 12, a focusing lens 13, etc. For example, the interferometer unit 17 can include a fiber collimator 7, a beam splitter 8, a focusing lens 10, a reflector 11, a controllable shutter 12, and a focusing lens 13, or alternatively, a fiber focuser 7, a beam splitter 8, a reflector 11, and a controllable shutter 12. In a typical example where a Michelson configuration (the main part of the measurement probe 18) with a beam splitter is used, low coherence illumination is collimated by the fiber collimator 7 and then split into a reference optical path and a sample optical path by the beam splitter 8. The beam in the reference optical path is partially reflected by the reflector 11 (such as a mirror or a retroreflective prism) and recombined with the return light in the sample arm that results from backscattering or reflection in the sample at the beam splitter 8. The interference between the resulting reflected reference optical path and the sample optical path light provides information about the Brownian dynamics of the NPs and can be used for the analysis of sample characteristics. The reference optical path can include an adjustable attenuator 9 that controls the reference beam intensity with respect to the sample beam intensity to optimize signal detection when in-process measurements are performed on a suspension flow where the particle size and optical / scattering characteristics are developing. The reference arm can also include a lens that focuses the reference beam onto a mirror 11, etc., or a dispersion compensating element or a focusing lens 10 that adapts to the dispersion generated in the sample arm.

[0028] The sample arm of the interferometer 17 can include means such as a controllable shutter 12 and a galvo - mirror that deflects light from the sample to enable measurement of only the light source spectrum returning from the optical and reference optical paths. Similarly, the attenuator 9 can also cancel out the reference beam so that only the spectrum of the scattered light from the sample is measured. These separate spectra can be periodically used for background removal in the primary data processing.

[0029] Furthermore, the sample arm may also include a lens 13 that focuses the beam into the sample. The focal position can be set to optimize the signal-to-noise ratio and may be controlled by a program, which can be executed, for example, by the control unit 5. The reference arm optical path length of the interferometer 17 can be set slightly shorter than the optical path length and up to the interface between the sample window 15 and the suspension 19.

[0030] Alternatively, the reference light having a fixed optical path length may also include the light reflected from the interface between the sample window 15 and the suspension or another fixed partially reflecting surface positioned between the lens 13 and the suspension 19. In such an arrangement, since both the generation and mixing of the reflected reference light and the scattered light from the sample are performed in the sample arm, the conventional separate reference arm, beam splitter 8, and shutter 12 can be omitted.

[0031] The optical fibers 16a to 16c can be several meters in length so that the components 1, 2, 4, and 5 can be arranged away from the measurement probe 18 in the process situation.

[0032] In this example, an interferometer unit using a Michelson interferometer is described, but other types of interferometer embodiments such as Mach-Zehnder can also be used.

[0033] The focus control and connection 14 and the sample window 15 can form an optical interface between the lens 13 and the suspension 19. This interface can realize a fixed connection between the focus control mechanism of the interferometer unit 17 and the transparent window 15 providing the suspension boundary. This alignment can be such that the optical axis of the sample arm is perpendicular to the direction of flow. The flow cell may include a temperature sensor 20 to provide the local temperature of the suspension in the flow device used for analysis.

[0034] Figure 2 shows an in-process implementation of an apparatus for non-invasive real-time suspension monitoring and process control in a continuous process. The figure shows a flow cell 201 (which can be implemented by a tube) through which a suspension 19 flows in the direction indicated by the arrow. The flow cell 201 is at least partially made of a transparent material or has a window made of a transparent material to enable the interferometer 17 to transmit light to the suspension 19 through the wall of the flow cell 201 and receive light from the suspension 19. Also shown in Figure 2 is that the control unit 5 can control the upstream process parameter 203 and the downstream process parameter 204.

[0035] In the implementation shown in Figure 2, the interferometer 17 may be coupled to an apparatus 21 such as a focus control device that includes the flow cell 201. The dimensions of the flow cell can be designed for a specific range of throughput to correspond to measurements within the accessible speed range for PSD analysis. The apparatus 21 may be included to adjust the depth of focus of the light emerging from the interferometer with respect to the flow cell 201. Furthermore, the angle φ between the optical axis and the inner surface of the cell can be set to 90 degrees or differently to avoid strong surface reflections. The flow cell may include a temperature sensor 20 to provide the local temperature of the suspension within the flow device for use in the analysis.

[0036] Figure 3 shows a schematic diagram of an apparatus for non-invasive real-time suspension monitoring and process control in a batch processing operation. As shown, the interferometer 17 is coupled to an adapter 304 of a reaction vessel 301 that includes a window 305 for optical access with means for inducing / aligning local flow in the form of grooves on the inner wall of the reaction vessel that are flush with the inner wall and allow the suspension to move in a layered manner in a direction controlled to enable measurement in strong turbulent flow. The temperature sensor 20 may be integrated with the window 305. The reaction vessel 301 includes a container as shown and may have means for agitation such as a stirrer. The stirrer may include a rotor blade 302 that can circulate the suspension 19 within the reaction vessel 301 in the direction indicated by the arrow 303 such that the flowing suspension is visible from the window 305. Means for setting the focus position and adjusting the angle between the window and the optical axis may also be provided.

[0037] In an embodiment similar to FIG. 3, the interferometer unit 17 and the window 305 to which at least one of the flow guiding device, the temperature sensor 20, and the adapter 304 can be attached can be implemented in a compact unit that forms a probe that can be inserted into the reaction vessel 301 to measure the suspension flow deeper inside the vessel away from the reaction vessel wall.

[0038] Different embodiments of the present device are used to achieve direct integration in a suspension synthesis / processing device and can provide 'real-time' particle size distribution data, for example, together with data analysis performed using parallel computing methods. Those skilled in the art will appreciate various alternative structures that also fall within the scope of the present invention.

[0039] Regarding the analysis of LCI data, the theory regarding the LCI signal in the presence of flow and Brownian diffusion of single-sized particles is disclosed in I Popov, A S Weatherbee, and I A Vitkin, “Dynamic light scattering arising from flowing Brownian particles: analytical model in optical coherence tomography conditions,” J Biomed Opt., vol 19, no 12, p 127004, 2014, and J Lee, W Wu, J Y Jiang, B Zhu, and D A Boas, “Dynamic light scattering optical coherence tomography,” Optics Express, vol 20, no 20 p 22262, 2012. This theory does not explain polydisperse suspensions. The present disclosure realizes the extension to polydisperse suspensions, which are the most important and common suspensions from an (industrial) application perspective.

[0040] In FDLCI, the signal I(t,λ) at the detector for each wavelength λ of the light source (i.e., the spectrum at time t) results from the interference of the electromagnetic field of the reflected reference beam and the light scattered from the population of NPs in the suspension. Due to the Brownian motion and other motions of the NPs, the signal I(t,λ) has temporal fluctuations characteristic of NP dynamics. The actual optical path length resolved signal I(t,z) of FDLCI is the complex Fourier transform of I(t,k), where k = 2π / λ and k is the wave number. In single scattering, a specific optical path length z represents a specific depth in the suspension, and the temporal fluctuations of I(t,z) reflect the NP motion in the coherence volume at that depth. In this case, for a suspension of NPs with a certain particle size distribution, the correlation function G(τ,z) = <|(0,z)| * (τ,z)> (where τ represents the delay time, '*' represents the complex conjugate, and <> represents the time average) can be shown to be accurately approximated by the following equation.

[0041] [Equation] and [Equation] Equation (1) assumes a low numerical aperture optical system (such as lens 13) and many non-interacting NPs in the coherence volume. The amplitude factor γ(z) is set by the detector characteristics of the optical system and the sample scattering characteristics, and g B (τ,z) reflects the Brownian motion and characterizes the local particle size distribution by the Stokes-Einstein relation and the right side of Equation (1). On the right side of Equation (1), Γ B = D r (4πn / λ c ) 2 is the decorrelation rate of the LCI signal scattered from the NPs with hydrodynamic radius r (D r is the diffusion coefficient, n is the solvent refractive index, and λ c is the central wavelength of the light source in vacuum). Furthermore, F z (ΓB ) is Γ B within the range [Γ B , Γ B + dΓ B , and thus is the ratio of the signal I(z) scattered from the NPs at the radius representing the PSD. K(Γ B τ, ξ) describes the decay of the autocorrelation of a single particle size. The form of the parameters α ξ , L p,ξ (p is an integer) depends on the coherence length ξ and can be derived from the results of [Popov]. For ξ > ~4 μm, α ξ , L ±1,ξ can be neglected to the extent that K(Γ B τ, ξ) = exp(−Γ B τ), and the right side of Equation (1) becomes the standard Laplace transform of F z (Γ B ).

[0042] In a suspension under flow, the FDLCI signal I(t, z) usually has additional temporal variations due to the average flow of the NPs that depends on the depth in the suspension. As calculations show, for the z-dependent flow velocities v x (z) perpendicular to the optical axis and v z (z) along the optical axis, G can be accurately approximated by the following equation.

[0043]

Equation

[0044] Alternatively, in a similar method where the modulus |I(t,z)| of I(t,z) is used, the corresponding intensity correlation G (2) (τ,z) = <|I(0,z)||I(τ,z)|> / <|I(z)|> 2 can be related to G(τ>0,z) in Equations (1) and (3) by the standard Siegert relation.

[0045] In principle, for the analysis of G(τ,z) at the suspension - confinement window boundary (where the flow v(z Bound ) = 0 holds and g Fx,z = 1, at z = z Bound ), the relevant Brownian contribution g B (τ) of G(τ,z), and thus the particle size and PSD scales, are already obtained. However, the signal - to - noise ratio can be significantly improved if information from z>z Bound (corrected for the influence of flow) is included. In any flow, the analysis of g B (τ,z) for a polydisperse suspension is very difficult, but in a frequently occurring quasi - stationary local laminar flow that can be orthogonal to the LCI beam, both the Brownian and cross - flow contributions to the autocorrelation can be extracted from G(τ,z) by Equation (3). In addition to the appropriate interfacial connection between the optical configuration and the flow, the present disclosure for particle size measurement in flow, for example, by Equations (3) and

Number

[0046] In a static suspension, when the optical path length (z - z Bound ) in the suspension exceeds the maximum optical path length Z ss characteristic of single scattering, the z-dependence of g B (τ,z) occurs due to multiple scattering. The characteristic optical path length Z ss related to the photon mean free path (MFP, typically Z ss ~5~10×MFP) is itself an important suspension characteristic and can thus be determined from the optical path length dependence of the measured FDLCI time fluctuations. In a flowing suspension, Z ss can be obtained from an iterative data analysis / fitting method, and thus <g B (τ)> is determined for z - z Bound <Z ss such that <g B (τ)> represents only single-scattered light and is thus guaranteed to be usable for the analysis of the average PSD (represented as <F(Γ B )>) by Equation (1). In practice, when z Bound ≒0, the autocorrelation g B (τ,z) is independent of the optical path length z when z < Z ss .

[0047] Figure 4 shows a flowchart of a method for obtaining and analyzing interference data. The overall method shown in the figure describes a continuous iterative cycle of obtaining and analyzing subsequent data blocks i for measurements in continuous mode during the monitoring of NP synthesis / processing operations. However, it is also possible to obtain and analyze a single data block to obtain information only once. As shown in Figure 4, at step 401, a time interval Δt iAfter obtaining a block of the FDLCI spectrum over a range, standard processing is performed in step 402, which may include performing an inverse Fourier transform and providing the optical path length resolved LCI scattering signal I(t, z) for each acquired time point t within the time interval and for each relevant optical path length z within the measured range of optical path lengths. In step 403, the z-resolved autocorrelation function G(τ, z), or the z-resolved frequency power spectrum

Number

[0048] In step 404, the G(τ, z) of the 'τ = 0 slice' γ(z), the noise level <G(τ -> ∞, z>, and the depth z * from which the signal disappears, as well as the statistical analysis of G(τ, z) evaluated over data sub-blocks within an interval Δt i to evaluate the stationary characteristics of the signal, can be performed. In other words, the amplitude and noise characterization of G(τ, z) are performed.

[0049] In step 405, an initial analysis of G(τ, z) for optical path lengths z within a narrow range of the suspension boundary at z Bound ) = 0 can be performed. This provides an initial estimate g Bound of the Brownian contribution to the autocorrelation G(τ, z). B,init (τ).

[0050] In step 406, the z-dependent flow-induced contribution g F,x (τ, Γ x (z)), and thus the fitting of the flow profile, can be performed, for example, using equation ( * ) with the spectrum G(τ, z < z * ) and the 'τ = 0 slice' γ(z < z 3 ) as inputs, setting g B (τ, z) = g B,init (τ) (initial spectrum). A delay time τ'(z) at which the residual exceeds the limit can be recorded. The obtained Γ x (z < z *By means of the low-order polynomial fitting of ( )), the analytical description v of the flow x fit (z < z * ) and the contribution g to the uncorrelated G(τ, z < z * ) due to the flow F,x fit (τ, z < z * ) of the associated analytical form can be obtained.

[0051] In step 407, the corrected form g of the Brownian part of the autocorrelation B cor (τ, z < z * ) is obtained from equation (3) using γ(z < z * ) and the analytical flow correction g F,x fit (t, z < z * ). The subsequent weighted partial averaging of, for example, g B (τ < τ’(z), z < z * ) within the determined delay time limit τ’(z) can provide an intermediate result of the average Brownian part <g B (τ)> of the autocorrelation function.

[0052] In step 408, the analysis of G fit (τ, z) = <g B (τ)> γ(z) g F,x fit (τ, z) and the residual ε(τ, z) = G(τ, z < z*) - G fit (τ, z < z*) can be performed. If it is determined that ε(z) exhibits systematic variations, the result g B cor (τ, z < z * ) can be considered depth-dependent (e.g., indicating multiple scattering). As shown in 408, in that case, for example, steps 406 to 408 can be repeated with a reduced z * = Z ss until depth independence is achieved at * .

[0053] In step 409, <g B(τ)>, the average particle size, polydispersity, and / or particle size distribution can be evaluated. This evaluation can be performed, for example, by cumulant analysis, CONTIN Laplace inverse transformation, or by equation (1) from <g B (τ)>, to extract information about <F(Γ B )> based on other known methods. Additionally, the number-based particle size distribution n(r) can also be extracted from the first-order distribution <F(Γ B )> using the standard Mie theory. The measured scattering length Z ss and the 'raw data' G(τ,z) or

Number

[0054] The above-described embodiments of the analysis method use the autocorrelation G(τ,z) of the FDLCI signal. However, alternatively, the average power spectrum of the Brownian contribution to the fluctuating FDLCI signal from it,

Number

Number

Number

Number

Number

[0055] The main purpose of the analysis of the present invention is the direct measurement of the particle size and PSD from G(τ,z) or [Number] However, other characteristics of G(τ,z) or [Number] such as the parameters that describe their time (frequency) and optical path length dependencies obtained by empirical fitting can also be used for the characterization of the flowing suspension for process monitoring or control.

[0056] Certain embodiments may include a computer program that can be used to control a computer, a control unit, or a processor to automatically control the acquisition settings of FDLCI devices, adapt these (such as reference intensity ratios, acquisition frequencies, etc.) in response to measurement signals, and monitor the time-dependent characteristics of a colloidal suspension. Additionally, the program may be configured to perform the described correlation analysis, data fitting, and obtain relevant characteristics. The program may incorporate parallel processing to perform these calculations in real time, which may mean that for the relevant processes, the analysis of data blocks is within the range of about 5 to 60 seconds. A control signal can be set to determine one or more relevant process parameters using the measured particle size or PSD information or other characteristics obtained from the LCI signal. In particular, the signal can be, for example, a trigger signal indicating that the average particle size or width of the distribution exceeds a certain limit.

[0057] The program can be implemented as computer-readable command code in any suitable programming language. The program elements can be stored on a computer-readable medium such as a CD ROM, DVD, Blu-ray disk, removable drive, volatile and non-volatile memory, etc. Further, the program elements may be provided over a network, such as the Internet, and downloaded on demand by the user.

[0058] The technology disclosed in this specification can be applied, for example, in a nanoprocessing process. The nanoprocessing process is a continuous processing process in which relatively large particles are processed into nanoparticles with a specified particle size and distribution. The in-line measurement system can be equipped with a flow cell adapter that can provide continuous real-time information regarding the particle size and distribution, using the measurements and calculations disclosed elsewhere in this disclosure. This information can be used as feedback or feedforward control to, for example, adjust the processing speed, stop the process, or continue if the target specifications have not (yet) been reached. In the absence of a possible method for in-line monitoring, the process can only be stopped to take a sample, or, if possible, draw a sample from a recirculation chamber and measure the sample offline. In such cases, the offline sample has to be analyzed externally and sample preparation is required before analysis, so that feedback to the process is significantly delayed. As the target specifications are approached, the process can be run non-stop with continuous monitoring data and stopped as necessary.

[0059] A second process control example is the monitoring of liposome formation for a pharmaceutical process by extrusion. Liposomes can be formed by passing through a specific lipid extrusion filter. Multiple extrusion cycles can be applied to obtain liposomes of similar size. By integrating the present invention into such a process, the actual particle size, polydispersity, and PSD can be evaluated in real time, and the process can be adjusted, continued, or stopped based on this information.

[0060] A third process control example where both the non-invasive nature and high-speed analysis capabilities of the present invention are suitably applied is the monitoring of a nanoparticle suspension as a final product within a vial or syringe during a filling and capping process. Measurement through the vial or syringe of the sample enables direct quality control of the final product. The data obtained can be used for process adjustment or for quality confirmation and real-time release of the final product.

[0061] Some or all aspects of the present invention may be suitable for implementation in the form of software, particularly computer program products. A computer program product may include a computer program stored on a non-transitory computer-readable medium. Also, the computer program may be represented by signals such as optical signals or electromagnetic signals carried by a transmission medium such as an optical fiber cable or air. The computer program may have a form of source code, object code, or pseudocode that is suitable for execution by a computer system, either partially or in whole. For example, the code may be executable by one or more processors. Such a processor is an example of a control unit.

[0062] The examples and embodiments described herein serve to illustrate rather than limit the present invention. Those skilled in the art will be able to design alternative embodiments without departing from the spirit and scope of the present disclosure as defined by the appended claims and their equivalents. The reference signs enclosed in parentheses in the claims are not to be construed as limiting the claims. Items described as separate entities in the claims or in this specification may be implemented as a single hardware or software item incorporating the features of the described items.

Examples

[0063] Next, the present invention will be described in more detail by several examples in which nanoparticle suspensions were characterized under different in-line conditions having different levels of flow. For this purpose, a suspension of calibrated polystyrene (PS) nanoparticles purchased from Kisker Biotech was used.

[0064] The first embodiment will be described with reference to FIGS. 5A and 5B. FIG. 5A shows the particle size distribution in a 1.25 mg / ml suspension of PS particles with a hydrodynamic radius of 48.5 nm, measured under stationary (0 rpm) and stirring conditions in a reaction vessel. The measurement configuration and optical system were prepared in the same manner as in FIG. 3 using a 15 cm diameter reaction vessel with a 4 mm thick glass window without a flow induction device, and the suspension was magnetically stirred at a speed of 50 - 250 rpm. The Reynolds number was in the range of Re ~ 400 - 1700 (from laminar flow to weak bulk turbulence). In FIG. 5A, the solid curve shows the particle size distribution at 0 rpm, the dashed curve shows the PSD under stirring conditions at a stirring speed of 100 rpm, and the dotted curve shows the PSD under stirring conditions at a stirring speed of 250 rpm. Data were obtained from the CONTIN analysis of the resulting <g B (τ)>. The data on suspension flow obtained from G(τ,z) using the disclosed analysis shows the laminar boundary layer within an optical path length of 1 mm in the suspension. The maximum shear rate within the layer corresponds to ~75 s at the maximum stirring speed of 250 rpm -1 . The data in FIG. 5B were measured at 4 - second intervals during a step - wise increase in the stirring speed within the container (shown in the upper panel) and show the average (Z - average) hydrodynamic radius (lower panel) analyzed in the manner described in the flow diagram of FIG. 4 using the cumulant analysis of <g B (τ). This analysis method, including flow correction, robustly characterizes the particle size and PSD over the range of stirring speeds used.

[0065] The second embodiment will be described with reference to FIG. 6. FIG. 6 shows the particle size distribution of a binary mixture of polystyrene particles with hydrodynamic radii of 49 nm / 191 nm at two concentration ratios of small particles to large particles, A: (1.25 mg / 0.07 mg) / ml and B: (1.25 mg / 0.22 mg) / ml, measured under flow in a flow configuration. The measurement configuration and optical system were prepared in the same manner as in FIG. 2 using a small flow cell (2 mm along the optical axis) carrying a flow of ~0.9 liters / hour generated by a constant pressurized upstream. In FIG. 6, the thick dashed line shows the particle size distribution at concentration ratio A, and the thin dashed line shows the corresponding cumulative distribution. The thick solid line shows the particle size distribution at concentration ratio B, and the thin solid line shows the corresponding cumulative distribution. For both concentration ratio A and concentration ratio B, independently, g F (τ,z) was analyzed to measure the flow velocity, and a Poiseuille flow profile with a peak velocity of ~23 mm / s and a maximum shear velocity of ~46 s -1 near the capillary / suspension boundary was obtained. The PSD of FIG. 6 as a result was obtained from the CONTIN analysis of <g B (τ)> for two different relative concentrations A and B. The data (peak width, peak position, and peak area ratio) are consistent with the expected distribution based on the particle sizes used, the individual polydispersities (coefficient of variation per species ~<7%), and the concentrations used, when considering the Mie theory to explain the scattering intensity.

[0066] Examples are disclosed in each of the following items.

[0067] 1. A method for monitoring the characteristics of colloidal particles in a flowing colloidal suspension, comprising: preparing a sample containing a flowing colloidal suspension; non-invasively monitoring at least one of the particle size and particle size distribution of the colloidal particles in the flowing colloidal suspension using Fourier domain, FD, low coherence interferometry, LCI, the monitoring step comprising deriving a decomposed LCI light scattering signal I(t,z) from a t-decomposed LCI wavelength spectrum of interference over time, t, and optical path length, z, a z-decomposed temporal autocorrelation function G(τ,z) of I(t,z) where τ represents a delay time, and a z-decomposed frequency power spectrum of I(t,z) where ω represents a frequency [Number] a step of deriving information indicating at least one of the particle size and the particle size distribution of the colloidal particles from at least one of A method comprising:

[0068] 2. From G(τ, z), determining the z- and τ-dependent uncorrelated factor g F (τ, z) related to the flow of the colloidal suspension; From G(τ, z), determining the z- and τ-dependent autocorrelation g B (τ, z) representing the Brownian motion of the colloidal particles; The method according to item 1, further comprising:

[0069] 3. From G(τ, z), the characteristic optical path length, Z, representing the photon mean free path in the flowing colloidal suspension, where g ss of z < Z B (τ, z) is independent of z within the measurement noise; ss determining; The method according to item 2, further comprising:

[0070] 4. Based on g ss of z < Z B (τ, z) in the flowing colloidal suspension, determining the average autocorrelation function, <g B (τ)>, representing the single-scattered light, and performing photon correlation spectroscopy, PCS, analysis to extract information regarding at least one of the particle size and the particle size distribution of the colloidal particles using <g B (τ)>; The method according to any one of items 1 to 3, further comprising:

[0071] 5. [Number] From [Number] and the associated z-decomposed power spectrum representing the Brownian motion-induced fluctuations [Number] the step of deriving [Number] from [Number] the step of deriving

[0072] 6. [Number] further comprising the step of deriving information related to at least one of the particle size and the particle size distribution from

[0073] 7. A step of performing FDLCI using low-coherence light having a central wavelength of about 300 nm to 3000 nm and using a swept light source method or a spectral region method, the swept light source method using a wavelength-swept light source, further comprising the step according to any one of items 1 to 6

[0074] 8. A step of flowing the colloidal suspension at a maximum speed of about 40 mm / s within an optical path length of about 3 mm from the measurement window or at a maximum shear rate of about 80 s within about 1 mm from the measurement window, further comprising the step according to any one of items 1 to 7 -1

[0075] 9. The method according to any one of items 1 to 8, wherein the particle size or the maximum dimension of the colloidal particles is about 10 nm to 3 μm, preferably about 15 nm to 1000 nm, and the method includes a step of determining at least one of the particle size and the particle size distribution within this particle size range.

[0076] 10. The method according to any one of items 1 to 9, wherein the colloidal suspension is a polydisperse suspension containing colloidal particles having at least two particle sizes within the range of about 10 nm to 3 μm, preferably about 15 nm to 1000 nm.

[0077] 11. The method according to any one of items 1 to 10, wherein the step of deriving I(t,z) and the step of obtaining information indicating at least one of the particle size and the particle size distribution of the colloidal particles based on I(t,z) are repeated periodically or continuously.

[0078] 12. The method according to any one of items 1 to 11, wherein the step of deriving information indicating at least one of the particle size and the particle size distribution of the colloidal particles is performed independently of the concentration of the colloidal particles in the suspension.

[0079] 13. The method according to any one of items 1 to 12, further including a step of controlling a process based on information indicating at least one of the determined particle size and the particle size distribution of the colloidal particles.

[0080] 14. The method according to any one of items 1 to 13, further including a step of classifying the suspension based on information regarding at least one of the particle size and the particle size distribution of the colloidal particles, and a step of controlling a step of classifying and monitoring the FDLCI device based on information indicating at least one of the particle size and the particle size distribution of the colloidal particles.

[0081] 15. A Fourier domain, FD, low coherence interferometry, LCI apparatus for monitoring the characteristics of colloidal particles in a flowing colloidal suspension in a process, A low coherence light source configured to simultaneously illuminate a reference optical path and a sample containing the flowing colloidal suspension, Means for obtaining an interference signal by mixing the light reflected from the reference optical path and the light scattered from the flowing suspension; A detector configured to detect a time-resolved LCI wavelength spectrum of interference between the light reflected from the reference optical path and the light scattered from a sample containing a flowing suspension; From the acquired LSI wavelength spectrum of interference, the time, t, and the optical path length, z, are derived, and the z-resolved temporal autocorrelation function G(τ,z) of the decomposed LCI light scattering signal, I(t,z), where τ is the delay time, or the frequency power spectrum of I(t,z), where ω is the frequency;

Equation

[0082] 16. A control unit configured to non-invasively monitor at least one of the particle size and the particle size distribution of colloidal particles in a flowing colloidal suspension, further configured to control at least one of the data acquisition and processing unit and at least one process parameter, the FDLCI apparatus according to item 15, further comprising the control unit.

[0083] 17. A computer-readable medium that, when executed by a computer, includes instructions to cause the computer to execute the method according to any one of items 1 to 14.

Explanation of Signs

[0084] 1... Light source, 2... Spectrometer, 4... Data acquisition and processing unit, 5... Control unit, 6... Direction element, 7... Fiber collimator or fiber focus, 8... Beam splitter, 9... Attenuator, 10... Dispersion correction or focusing lens, 11... Reflector, 12... Controllable shutter, 13... Focusing lens, 15... Sample window, 17... Interferometer unit, 18... Measurement probe, 19... Suspension.

Claims

1. A suspension containing nanoparticles, which is a method for determining the characteristics of nanoparticles in a laminar, quasi-steady flowing suspension, comprising: preparing a sample containing a suspension containing nanoparticles; non-invasively determining at least one of the particle size and particle size distribution of the nanoparticles in the suspension using Fourier domain low coherence interferometry (FD-LCI), wherein a low coherence beam is perpendicular to the flow direction of the suspension, and the determining step comprises: (a) deriving a time (t) and optical path length (z) resolved LCI light scattering signal I(t, z) from the t-resolved LCI wavelength spectrum of interference; and (b) deriving information indicating at least one of the particle size and the particle size distribution of the nanoparticles from the z-resolved temporal autocorrelation function G(τ, z) of I(t, z), where τ represents the delay time; including; the step (b) comprising: From G(τ, z), a z- and τ-dependent decorrelation factor g related to the flow of the suspension Ｆ (τ, z) is determined, wherein the decorrelation factor g Ｆ (τ, z) describes the time variation of I(t, z) induced by the flow of the suspension, and first, the amplitude and noise characteristics of G(τ, z) are characterized, and second, at a z Ｂｏｕｎｄ at which the zero velocity condition v(z = z Ｂｏｕｎｄ ) = 0 holds, an initial analysis of G(τ, z) for the optical path length z within the range of the suspension boundary is performed to provide an initial estimate g Ｂ，ｉｎｉｔ (τ) of the Brownian contribution to the autocorrelation function G(τ, z), and third, a fit of the z-dependent flow-induced contribution g Ｆ (τ, z) in G(τ, z) is performed, which is executed by Determined g Ｆ Using (τ, z), from G(τ, z), determine the z- and τ-dependent autocorrelation g representing the Brownian motion of the nanoparticles Ｂ Step of determining (τ, z), From G(τ,z), for z < Z in the suspension ｓｓ of g Ｂ (τ,z) is independent of z within the measurement noise, and determining a characteristic optical path length (Z ｓｓ ) that represents the photon mean free path in the suspension; z < Z in the suspension ｓｓ of g Ｂ Based on g(τ, z), determine the mean autocorrelation function <g Ｂ (τ)> representing single scattered light, and use <g Ｂ (τ)> to perform photon correlation spectroscopy (PCS) analysis to extract the information regarding at least one of the particle size and the particle size distribution of the nanoparticles a method.

2. A suspension containing nanoparticles, which is a method for determining the characteristics of nanoparticles in a laminar, quasi-steady flowing suspension, comprising: preparing a sample containing a suspension containing nanoparticles; non-invasively determining at least one of the particle size and particle size distribution of the nanoparticles in the suspension using Fourier domain low coherence interferometry (FD-LCI), wherein a low coherence beam is perpendicular to the flow direction of the suspension, and the determining step comprises: (a) deriving a time (t) and optical path length (z) resolved LCI light scattering signal I(t, z) from the t-resolved LCI wavelength spectrum of interference; and (b) deriving information indicating at least one of the particle size and the particle size distribution of the nanoparticles from the z-resolved temporal frequency power spectrum of I(t, z), where ω represents the frequency; 【Number 1】 including; the step (b) comprising: performing an initial analysis of the z-resolved power spectrum 【Number 2】 related to the flow of the suspension to provide an initial estimate 【Number 3】 A step of determining, firstly, the optical path length z within the range of the suspension boundary at z where the zero velocity condition v(z = z Ｂｏｕｎｄ ) = 0 holds Ｂｏｕｎｄ ​ 【Number 4】 of the Brownian contribution to the z-resolved temporal frequency power spectrum; 【Number 5】 thirdly, 【Number 6】 performing an inverse convolution of 【Number 7】 from 【Number 8】 to determine the z-resolved power spectrum representing the Brownian motion of the nanoparticles; 【Number 9】 determining from 【Number 10】 the average power spectrum representing single scattered light; 【Number 11】 From, in the suspension, z < Z ｓｓ of 【Number 12】 Determining a characteristic optical path length (Z ｓｓ ), which represents the photon mean free path in the suspension and is independent of z within the measurement noise; z < Z in the suspension ｓｓ of 【Number 13】 and determining from 【Number 14】 information related to at least one of the particle size and the particle size distribution of the nanoparticles; 【Number 15】 including; a method.

3. A step of performing the FDCLI using a low-coherence light having a central wavelength of 300 nm to 3000 nm and using a swept light source method or a spectral region method, the swept light source method using a wavelength-swept light source, further including the step, the method according to claim 1 or 2.

4. Flowing the suspension at a maximum speed of 40 mm / s within an optical path length of 3 mm from the measurement window, or at a maximum shear rate of 80 s within 1 mm from the measurement window. －１ The method according to any one of claims 1 to 3, further comprising the step of flowing at a shear rate of.

5. The method according to any one of claims 1 to 4, wherein the particle size or the maximum dimension of the nanoparticles is 10 nm to 3 μm, and the method includes a step of determining at least one of the particle size and the particle size distribution within this particle size range.

6. The method according to any one of claims 1 to 5, wherein the suspension is a polydisperse suspension containing nanoparticles having at least two particle sizes within the range of 10 nm to 3 μm.

7. The step of deriving I(t, z) and the step of obtaining the information indicating at least one of the particle size and the particle size distribution of the nanoparticles based on I(t, z) are repeatedly performed periodically or continuously, the method according to any one of claims 1 to 6.

8. The method according to any one of claims 1 to 7, wherein the step of deriving the information indicating at least one of the particle size and the particle size distribution of the nanoparticles is performed independently of the concentration of the nanoparticles in the suspension.

9. The method according to any one of claims 1 to 8, further including a step of controlling a process based on the determined information indicating at least one of the particle size and the particle size distribution of the nanoparticles.

10. The method according to any one of claims 1 to 9, further including a step of classifying a suspension based on the information regarding at least one of the particle size and the particle size distribution of the nanoparticles, and a step of controlling an FDCLI device and the step of determining based on the information indicating at least one of the particle size and the particle size distribution of the nanoparticles.

11. A suspension containing nanoparticles, a Fourier domain low-coherence interferometry (FDCLI) device for determining the characteristics of the nanoparticles in a suspension flowing in a laminar quasi-steady state in a process, A low-coherence light source configured to simultaneously illuminate a reference optical path and a sample containing the suspension, the low-coherence light source being orthogonal to the flow direction of the suspension, and a low-coherence beam reflected from the sample, means for obtaining an interference signal by mixing light reflected from the reference optical path and light scattered from the suspension; a detector configured to detect a time-resolved LCI wavelength spectrum of interference between the light reflected from the reference optical path and the light scattered from the sample including the suspension; From the obtained LCI wavelength spectrum of the interference, the time (t) and optical path length (z) resolved LCI optical scattering signal I(t, z) is derived, and from the z-resolved temporal autocorrelation function G(τ, z) of I(t, z), where τ is the delay time, information indicating at least one of the particle size and the particle size distribution of the nanoparticles is derived. From G(τ, z), the z- and τ-dependent decorrelation factor g Ｆ (τ, z) that describes the temporal variation of I(t, z) induced by the flow of the suspension is determined. The decorrelation factor g Ｆ (τ, z) is determined by, first, characterizing the amplitude and noise of G(τ, z), second, performing an initial analysis of G(τ, z) for the optical path length z within the range of the suspension boundary at z = zBound where the zero velocity condition v(z = zBound) = 0 holds, to provide an initial estimate gB,init(τ) of the Brownian contribution to the autocorrelation function G(τ, z), and third, fitting the z-dependent flow-induced contribution gF(τ, z) in G(τ, z). Using the determined g Ｆ (τ, z), the z- and τ-dependent autocorrelation, g Ｂ (τ, z), representing the Brownian motion of the nanoparticles is determined from G(τ, z), and from G(τ, z), for z < Z ｓｓ in the suspension, the g Ｂ (τ, z) is independent of z within the measurement noise, and the characteristic optical path length (Z ｓｓ ) representing the photon mean free path in the suspension is determined. A data acquisition and processing unit configured to derive the information indicating at least one of the particle size and the particle size distribution of the nanoparticles, based on g ｓｓ (τ, z) for z < Z Ｂ in the suspension, the mean autocorrelation function <g Ｂ (τ)> representing single-scattered light is determined, and further configured to perform photon correlation spectroscopy (PCS) analysis to extract the information regarding at least one of the particle size and the particle size distribution of the nanoparticles using <g Ｂ (τ)>. An FDLC I apparatus comprising the above, wherein the low coherence light source is a broadband light source or a swept light source, and the swept light source is a wavelength-swept light source.

12. A suspension containing nanoparticles, which is a Fourier domain low coherence interferometry (FDLC I) apparatus for determining the characteristics of nanoparticles in a flowing suspension in a process, A low coherence light source configured to simultaneously illuminate a reference optical path and a sample including the suspension, wherein the low coherence beam reflected from the sample is orthogonal to the flow direction of the suspension; means for obtaining an interference signal by mixing light reflected from the reference optical path and light scattered from the suspension; a detector configured to detect a time-resolved LCI wavelength spectrum of interference between the light reflected from the reference optical path and the light scattered from the sample including the suspension; Deriving a time (t) and optical path length (z) resolved LCI light scattering signal I(t, z) from the acquired LCI wavelength spectrum of interference, where ω represents frequency, and the z-resolved temporal frequency power spectrum of I(t, z); 【Number 16】 Deriving information indicating at least one of the particle size and particle size distribution of the nanoparticles from; 【Number 17】 From the z-resolved power spectrum related to the flow of the suspension 【Number 18】 To First, the optical path length z within the range of the suspension boundary at z where the zero velocity condition v(z = z Ｂｏｕｎｄ ) = 0 applies Ｂｏｕｎｄ ​ 【Number 19】 Perform an initial analysis of to obtain the z-resolved temporal frequency power spectrum 【Number 20】 An initial estimate of the Brownian contribution to 【Number 21】 To provide, thirdly, 【Number 22】 From 【No. 23】 Determined by inverse convolution of to obtain 【24 Points】 From, the z-resolved power spectrum representing the Brownian motion of the nanoparticles 【Number 25】 To determine 【Number 26】 From, z < Z in the suspension ｓｓ of 【Number 27】 The characteristic optical path length (Z ｓｓ ), which represents the photon mean free path in the suspension and is independent of z within the measurement noise, is determined by a data acquisition and processing unit configured to derive the information indicating at least one of the particle size and the particle size distribution of the nanoparticles, for z < Z ｓｓ in the suspension 【Number 28】 Based on, the average power spectrum representing single-scattered light 【Number 29】 To determine 【30 numbers】 A data acquisition and processing unit further configured to derive information related to at least one of the particle size and the particle size distribution of the nanoparticles from; An FDLC I apparatus comprising the above, wherein the low coherence light source is a broadband light source or a swept light source, and the swept light source is a wavelength-swept light source.

13. A control unit configured to non-invasively monitor at least one of the particle size and the particle size distribution of the nanoparticles in the suspension, the control unit being further configured to control at least one of the data acquisition and processing unit and at least one of the at least one process parameter, the FDLC device according to claim 11 or 12, further comprising a control unit.

14. A computer-readable medium including instructions that, when executed by a computer, cause the computer to execute the method according to any one of claims 1 to 10.

Images