Keywords

1 Introduction

Over the last decades, nanomaterials (NMs) have become extremely popular thanks to unique properties that can be exploited in different fields such as energy (Ravi and Vadukumpully 2016; Dessie et al. 2019) , transportation (Jung et al. 2005; Ali et al. 2018) , industry (Khalil et al. 2017; Jørgensen 2009) , food (Dubascoux and Wyser 2019) , cosmetics (Katz et al. 2015) , and medicine (Han et al. 2019; Abd Elkodous et al. 2019) . They can occur with different structures and be composed of various matter such as metals , that is, titanium oxide, gold, silver, platinum, and ferric oxides, polymers, lipids, carbons including carbon nanotubes, graphene derivatives, nanodiamonds, and fullerenes.

Many NMs have found interest in medical applications . Pharmaceuticals and medical devices based on the use of these technologies were called nanomedicines . They include various types of nano-objects which vary in their structure and composition. It was a rapidly growing field over the past two decades but several aspects on their definition remain under debate. There is a need to clarify the classification of the different types of nanomedicines occurring with complex structures (Castagnola et al. 2017). Regarding the size , the definition given for a NM proposed by authorities in early 2010 is too narrow to include all types of nanomedicines as it excludes many nanomedicines whose size is larger (200–300 nm) than the upper limit given in the official definition based on at least one dimension lower than 100 nm for 50% of the number size distribution of NMs. Nevertheless, a consensus is established on the need to provide with relevant quality control procedures to assess product quality insuring repeatability and reproducibility of the safety and efficacy on a batch-to-batch basis. This can be achieved performing the characterization of NMs by the use of validated procedures under conditions compatible with quality control (Varenne et al. 2015a, b; Loeschner et al. 2015; Linsinger et al. 2013; Dudkiewicz et al. 2015; Braun et al. 2011a) or methods whose performances have been proven by interlaboratory comparisons (Linsinger et al. 2014; Weigel et al. 2017; Lamberty et al. 2011) thus ensuring reliable results. The reliability of measurements can be ensured by defining a series of handling precautions and quality criteria for good measurements (Varenne et al. 2015a, b, c, d). The selection of relevant methods to characterize properties of NMs should be performed by comparing available methods to provide reliable measurements (Varenne et al. 2016a; Till et al. 2016; Teulon et al. 2018; Sokolova et al. 2011; Grombe et al. 2014; Cascio et al. 2014; Anderson et al. 2013; Sikora et al. 2015; Borchert et al. 2005; Aichele et al. 2015). It is noteworthy that the characterization of physicochemical parameters of NMs in general remains a difficult task even for parameters including the size of the nano-object and the distribution of size, the surface charge using automatic measurement instruments. Most characterization methods of NMs require a preparation of the sample that will be used to perform measurements with the specifically designed method. This can include a dilution of the sample or the realization of a dry depot on a substrate. Whatever the modalities for the preparation of the sample, efforts are needed to ensure that measurements will be representative of the original dispersions of NMs (Varenne et al. 2015a, b, c, d; Ghomrasni et al. 2020; Wagner et al. 2015; Delvallée et al. 2015). This chapter aims to give some practical guidelines to characterize nanomedicine-based pharmaceuticals in the quality control assessment perspective.

2 Characterization of Materials

The characterization of NMs under conditions compatible with quality control is a societal task. NMs are characterized by two different types of parameters. For instance, the composition, the concentration, the structure, and the surface functionalization of the NMs are general parameters which are not restricted to NMs, although methods for the determination of the concentration are very specific. Specific characteristics of NMs include their size parameters, giving the size , the particle size distribution (PSD) , and the agglomeration or aggregation state , their surface properties as surface charge through the evaluation of the zeta potential (ZP) , reactive surface, surface area and porosity, and their shape (Hassellöv and Kaegi 2009; Guidance manual for the testing of manufactured nanomaterials 2010). These characteristics should be characterized as suggested by the technical committee of International Organization for Standardization (ISO TC 229 – Nanotechnologies) and the OECD Working Party on Manufactured Nanomaterials. Modifications of size parameters and surface properties of nanomedicines can affect their biological fate hence biological efficacy and safety (Shekunov et al. 2007; Gaumet et al. 2008). Size parameters and surface properties of NMs are among paramount factors to evaluate in order to assess repeatability and reproducibility and efficiency of industrial processes and product quality (Li 2010). Table 23.1 summarizes the different methods that are available to assess specific physicochemical parameters of NMs. It points out direct and indirect methods and those that can be applied in routine analysis. Table 23.2 overviews the general physicochemical parameters that are used to describe the properties of NMs. It highlights the methods that can be applied to assess these general parameters. It is noteworthy that the evaluation of the concentration of NMs can be performed using the methods specific to the NMs. The application of any mentioned method of characterization in quality control analysis needs to be validated according to general procedures used in metrology in order to provide uncertainties associated to the measurement of the physicochemical parameter of NMs using a given method and applying a specific measurement procedure .

Table 23.1 Specific physicochemical parameters used to describe properties of NMs (Hassellöv and Kaegi 2009; Linsinger et al. 2012)
Full size table
Table 23.2 General parameters used to describe properties of NMs
Full size table

3 General Consideration to Achieve Quality Control Analysis and Metrology: Validation and Transfer of Analytical Procedures

The characterization of NMs is necessary to describe the properties of the NMs composing nanomedicines thus achieving safety-efficiency and batch-to-batch consistency. In practice, very few methods are available to achieve the characterization of nanomedicines on a routine basis considerably limiting the number of parameters that can be included in quality control assessment . It is noteworthy that almost all methods are indirect methods, which means that the parameter measured by the instrument is then used to calculate the property desired to determine. Models developed to convert the measure into the measurand can be quite complexed, restricting the application of the technique to the characterization of a narrow range of NMs. Standardization of size measurement methods is paramount to provide results that are comparable between laboratories. For instance, the widely used method for size determination by dynamic light scattering (DLS) can be applied on spherical NMs with a narrow size distribution . Results are biased while the polydispersity increases, and the method is inappropriate to characterize the size of non-spherical particles. Measurements should be performed under conditions compatible with quality control that requires the use of standardized procedures. The procedures should be validated and uncertainties should be evaluated with a reference NM close to that which will be analyzed. Moreover, instruments must be qualified using appropriate reference NMs including materials from National Institute of Standards and Technology (NIST) when available. In the quality control assessment procedure for the analysis of a NM, the reference NMs should be analyzed before and after the analysis of “unknown” NMs by the same validated measurement procedure .

The evaluation of physicochemical properties of NMs should be performed under conditions which are compatible with quality control to provide reliable characterization. Reliability of results can be appreciated with associated measurement uncertainty determined through the validation of analytical procedures. The validation of analytical procedures consists in providing guarantees with certified reference material (CRM) or reference material (RM) that analytical procedures are sufficiently acceptable, reliable, and adequate for elements of their scope (ICH 1994; ISO 5725-1 1994; Ahuja and Scypinski 2001). Moreover, laboratories should prove that their analysts are able to perform analytical procedures with similar results (Ahuja and Scypinski 2001; Code of Federal Regulations; USP 37, General Information 1224). Hence, there is a need to provide guidelines to ensure quality control and thereby to evaluate the safety and toxicity of NMs. Draft guidance documents are provided for manufactured NMs, indicating various methods that can be applied to evaluate these parameters (Guidance manual for the testing of manufactured nanomaterials 2010; ISO/TS 80004-6 2013). Nevertheless, no indication is given to validate and transfer analytical procedures applied to the characterization of NMs and to provide uncertainty closed to results (Guidance manual for the testing of manufactured nanomaterials 2010).

Although many parameters can be used to define one NM, only a few are really accessible for a routine analysis using marketed instruments or having been the subject of standards from International Organization for Standardization (ISO) description as size (ISO 13318-3 2004; ISO 13318-2 2007; ISO 13318-1 2001; ISO 22412 2017; ISO/TS 21362 2018; ISO 13321 1996; ISO 29301 2017; ISO/DIS 21363; ISO/DIS 19749; ISO 13322-1 2014; ISO/TS 13762 2001; ISO 11039 2012; ISO 27911 2011; ISO 20998-1 2006; ISO 20998-2 2013; ISO 20998-3 2017; ISO/DIS 15900) , surface charge (ISO 13099-3 2012; ISO 13099-2 2012; ISO 13099-1 2012), shape (ISO/DIS 21363; ISO/DIS 19749; ISO/TS 10797 2012), surface area (ISO 18852 2012; ISO 18757 2003), and reactive surface (ISO/AWI TS 23459).

3.1 Validation and Transfer of Analytical Procedures

Whatever the type of analysis , it follows a well-established analytical procedure describing in detail all steps needed to carry out a given analysis. All analytical procedures will follow a life cycle which includes a validation stage and a transfer stage as illustrated in Fig. 23.1. The validation is achieved applying strict metrology concepts which aim to prove that the analytical procedure is sufficiently acceptable, reliable, and adequate for the elements of its scope (ICH 1994; ISO 5725-1 1994; Ahuja and Scypinski 2001). The validation is generally achieved using CRM or RM . It consists of performing numerous measurements of these materials following the described procedure. The results are then analyzed with appropriate statistical analytical methods. The guide to the expression of uncertainty in measurement (GUM) outlines statistical methodologies to interpret raw data of validation as analysis of variance (ANOVA) (Evaluation of measurement data – guide to the expression of uncertainty in measurement 2008) . The different parameters evaluating performances of analytical procedures were summarized in Table 23.3. The validation of analytical procedure permits to assess to the associated expanded uncertainty expressing reliability of results provided with validated analytical procedure (Evaluation of measurement data - guide to the expression of uncertainty in measurement. 2008). CRM is a material that is metrologically characterized with valid procedure for one or more specified properties (ISO Guide 35 2006). Analysis certificate providing value of specified property with corresponding uncertainty and metrological traceability is produced with CRM. RM is a homogeneous and stable material toward one or more specified properties (ISO Guide 35 2006). It is adequate for its used in process of measurement of specified property. When it is possible, it is important to validate the analytical procedure with a material certified for the analytical method that will be used. The number of CRM and RM available to validate methods of characterization of NMs is limited. Size CRM generally consists in monodispersed NMs. Only one consists in bimodal dispersion of silica nanoparticles (NPs) certified at 18.2 and 84 nm with electron microscopy (EM) (ERM-FD102). There is only one available CRM with assigned SI-traceable values of positive electrophoretic mobility (NIST Standard Reference Material® 1980, value: 2.53 ± 0.12 μm.cm.V−1.s−1). It is noteworthy that there is another CRM with a negative value of ZP (ERM-FD100, value: 43.0 ± 21.8 mV (Braun et al. 2011b)). However, the uncertainty of the certified value of ZP of this standard is about 50% of the certified value. Other CRMs are currently under development (Levin et al. 2018). Polystyrene latex particles–based standard is commercially available but it is not a CRM (DTS1235 from Malvern , value: 42.0 ± 4.2 mV).

Fig. 23.1
figure 1

Life cycle of analytical procedure

Full size image
Table 23.3 Overview of parameters used to describe the performance of analytical procedures (ICH 1994; ISO 5725-1 1994; Ahuja and Scypinski 2001)
Full size table

Besides having appropriate CRM or RM , validation also needs to investigate adequate parameters. No official specific guidelines were yet established to perform the validation of a measurement procedure characterizing NMs. The guidelines Q2(R1) from International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH guidelines Q2(R1)) was established only for the validation of most common types of analytical procedures including identification tests, quantitative tests for impurities’ content, limit tests for the control of impurities, and quantitative tests of the active moiety in samples of drug substance or drug product or other selected components in the drug product (ICH 1994). Other types of analytical procedures such as dissolution testing of drug products and the evaluation of particle size of drug substance have not been addressed in this document. This guideline mentioned that the validation of these analytical procedures is equally important to those listed herein and may be considered in subsequent documents. Although this guideline did not provide any specific information on how validation of NM characterization procedures should be carried out, concepts to achieve such validations can be drawn from it. The selection of studied parameters should be adapted on a case-by-case basis.

Other official documents propose some lines to perform validation of measurement procedure s applicable to the characterization of NMs. Standards from ISO suggest to study trueness and precision , that is, repeatability and intermediate precision of procedures used to evaluate the ZP of NMs with ELS coupled to phase analysis light scattering (PALS) (ISO 13099-2 2012) and precision, that is, repeatability and reproducibility of procedures applied to evaluate the size of NMs by DLS (ISO 22412 2017). However, no indication about the number of samples needed to study each parameter and statistical methodologies to interpret raw data was given in ISO standards (ISO 22412 2017; ISO 13099-2 2012). The Nanomedicine Characterization Laboratory (Frederick, MD, USA) proposes standardized procedures to evaluate the size of NMs with DLS (Hackley and Clogston 2007), atomic force microscopy (AFM) (Grobelny et al. 2009), transmission electron microscopy (TEM) (Bonevich and Haller 2010) , scanning electron microscopy (SEM) (Vladár and Ming 2011) , and electrospray-differential mobility analysis (ES-DMA) (Pease III et al. 2010) or to evaluate ZP (Clogston 2009). It was reported that procedures used for evaluating the size of NMs by DLS (Hackley and Clogston 2007) and procedures applied to size evaluation of NMs with SEM (Vladár and Ming 2011) should be validated . Last decade, Shekunov et al. and Gaumet et al. were the first to carry out reflexion about the reliability of results for NM characterization through size measurement with acceptable trueness (Shekunov et al. 2007; Gaumet et al. 2008) .

A measurement procedure validated in one laboratory can be transferred to other laboratories through a transfer approach. The aim is to demonstrate that the procedure validated by the sending laboratory can be applied in the other laboratories, named receiving laboratories, with the same performances. It must prove that receiving laboratories are able to carry out analytical procedure by providing similar results as the sending laboratory (Ahuja and Scypinski 2001; Code of Federal Regulations; USP 37, General Information 1224). Approaches that can be used to achieve the transfer of an analytical procedure are described by the Food and Drug Administration (FDA) (Ahuja and Scypinski 2001) and the USP Pharmacopeia (USP 37, General Information 1224) . They are also described in the Handbook of Modern Pharmaceutical Analysis (Ahuja and Scypinski 2001) . The different approaches that can be included in a transfer of analytical procedure are summarized in Table 23.4. Their selection to achieve the transfer of a given analytical procedure depends on risk assessment , complexity, criticality, and the aim of the analytical procedure. In general, during the analytical stage, each laboratory including the sending laboratory and all receiving laboratories analyze the same batch of samples. Data obtained from the different laboratories are compared and confronted to acceptance criteria that are defined depending on the method. It is noteworthy that no specific information is provided to perform the transfer of physicochemical characterization procedures of NMs. The selection of a suitable approach to transfer such a procedure should be adapted on a case-to-case basis.

Table 23.4 Overview of approaches used for the transfer of analytical procedures (Ahuja and Scypinski 2001; Code of Federal Regulations; USP 37, General Information 1224)
Full size table

3.2 Qualification of Instrument

The qualification of an instrument is achieved to provide documented evidence that the instrument performs with specification. According to the ISO standard and the Good Manufacturing Practices , the instruments should be calibrated or checked by appropriate methods with suitable control samples as traceably calibrated materials at defined periods (ISO 9001 2008; Good manufacturing practices). There are different stages of qualification covering the life of an instrument from its design to its utilization in routine (Fig. 23.2).

Fig. 23.2
figure 2

Stages of qualification of an instrument

Full size image

This aspect was introduced in the ISO standard devoted to the measurement of size of NMs by DLS (ISO 22412 2017). The ISO standard mentions that the qualification of the instrument should be performed after installation (operational qualification) and at regular time intervals (performance qualification) with a dispersion of materials with certified size. CRM with values assigned for DLS using the same algorithm to determine the size of the CRM should be used to carry out the qualification of the instrument. It is mentioned that the chemistry and the morphology of the NMs constituting the CRM should match the test samples as closely as possible. It is noteworthy that, alternatively, certified dispersions of polystyrene latex with narrow size distribution with average particle diameter as evaluated by DLS or EM can be used for the qualification of instrument. The qualification of the instrument can be evaluated either from five repeat measurements of size of CRM by comparing the difference between the measured average and the certified values and the expanded uncertainty closed to the measured average value (Linsinger 2005) or from three repeat measurements of size of CRM carried out before and after the measurement of the size of unknown NMs; the size of the CRM should be within the range of size determined during the validation of the procedure used to evaluate the size of unknown NMs (Varenne et al. 2015b, 2016b). If the qualification fails, it can indicate a mistake in the preparation of the dispersion or the instability of the dispersion or the failure of the instrument .

4 Validation of Procedures Evaluating Physicochemical Parameters of NMs: Examples

4.1 Size Measurement by Dynamic Light Scattering

DLS is a major technique used to measure the size of NMs. This method is very popular thanks to the existence of easy to use affordable marketed measurement instruments. DLS was also implemented to achieve continuous measurements using a glass capillary mounted in classical laboratory instrument (Ruseva et al. 2018). Results provided with this method are reliable considering NMs of homogenously distributed size have a narrow size distribution (Varenne et al. 2015b, 2016b) . However, this technique should be applied with caution when characterizing the size of unknown NMs as bias on measurements can be introduced in the case of non-homogenous in size dispersions or of dispersions showing a wide or complex polydispersity (Varenne et al. 2016a; Sokolova et al. 2011; Cascio et al. 2014; Anderson et al. 2013; Langevin et al. 2018a; Elizalde et al. 2000).

Very few works have reported size results with associated measurement uncertainty ensuring reliable characterization of size of NMs by DLS (Varenne et al. 2015b, 2016b, 2019b; Braun et al. 2011a). The preparation of the sample to perform size measurement by DLS is particularly a critical step (Varenne et al. 2015b, d, 2016b; Braun et al. 2011a; Langevin et al. 2018a, b). The presence of dust may compromise the size measurement of NMs. It is necessary to prepare diluted samples of NMs with freshly filtered dispersants with 0.22 μm filter and flasks with caps should be pre-rinsed with filtered ultrapure water and stored in a dust-free environment. Bias can be introduced with the quality of measurement macrocuvettes. Cuvettes showing defects on the optical faces must be discarded while they can represent 85% of the units in a box depending on suppliers and quality. The measurement cuvettes should be cleaned with filtered ultrapure water and stored in a dust-free environment until use. The measurement cuvettes should be used only once to avoid cross-contamination. The volume of sample introduced in the macrocuvette should be sufficient to permit the passage of the laser into the sample. The larger the volume is, the longer the equilibration time is to let the sample reach the temperature of measurement. Indeed, the temperature of the sample during measurement is paramount to control to provide with reliable size results as the measured parameter is the diffusion coefficient from which the size is calculated using the Stokes and Einstein equation . Artifacts due to degassing of the samples may be created with high difference between the temperature of the sample and the temperature of measurement. The equilibration time should be long enough for the sample to achieve the temperature of measurement. A minimum of 1 min per degree of difference should be considered for a volume sample of approximately 1 mL. Optimal concentration of the dispersions of NMs to carry out size measurement should be optimized for the intensity of the signal to be within the range recommended by the supplier of the instrument used. For this purpose, the curve representing the intensity of the signal as a function of the concentration of NMs should be established and the optimal concentration is selected on the linear part of this curve (Cao 2003).

Some quality criteria should be defined and followed to ensure reliable results (Varenne et al. 2015b, d). For size measurement by DLS, the quality of the correlogram reflecting the probability to find the NMs at the same place after a few times and the count rate curve corresponding to the number of photons collected by the detector associated to each run during measurements can be followed during the size measurement. After measurement, the raw correlogram, the intercept describing the amplitude of the correlogram that is close to the signal-to-noise ratio, the mean count rate, and the cumulant fit error can be inspected. The cumulant fit error is the closeness of agreement between the experimental raw correlogram and the calculated correlogram by means of the cumulant method described in the ISO standard (ISO 22412 2017).

It is noteworthy that an ISO standard dealing with good practice for DLS measurements is under development (ISO/PRF TR 22814).

The selection of the CRM or RM is crucial. NIST Traceable Particle Size Standards consisting in polystyrene latex standard with SI-traceable certified values by TEM can be used to validate the developed procedures. These CRM are spherical NPs known to not swell in aqueous dispersions and appeared quite monodisperse as acknowledged by the low PDI (PDI < 0.05) and available from 50 to 900 nm. These CRMs should be diluted in NaCl 10 mM for suppressing the electrical double layer and ensuring that the measured hydrodynamic diameter was the same as expected by TEM as described in the ISO standard (ISO 13321 1996). Other CRM with traceable mean diameter of 20, 30, and 40 nm by DLS are available.

The developed procedures should be validated by studying robustness, precision , that is, repeatability and intermediate precision, and trueness to evaluate the expanded uncertainties of the procedures. The robustness is investigated by varying experimental parameters that may influence measurements of size of NMs permitting to provide indication on the reliability under normal conditions of use of the proposed procedures. This study is a preliminary step before transferring methods to other laboratories or performing collaborative studies. The repeatability is performed by measuring the size of the CRM carried out successively in the same day and the intermediate precision by measuring the size of the CRM performed in different days. In the experimental nested design proposed by Varenne et al., to investigate the precision of the procedure, three samples of diluted CRM at optimal concentration were analyzed per day (Varenne et al. 2015b). Each sample was analyzed in triplicate, that is, three successive size measurements were performed on each sample. This experimental nested design permits to investigate the influence of the factors days, samples, and replicates that are considered as random (Fig. 23.3). The raw data were interpreted by means of ANOVA permitting to investigate the variability between days, between samples variability analyzed on the same day (within days), and between replicates variability of a sample (within samples). Appropriate statistical models were developed to interpret the raw data. According to the ISO standard (ISO 22412 2017), the relative uncertainties of repeatability and reproducibility should be below 2% and 5%, respectively. This ISO standard mentions any information about the evaluation of intermediate precision to evaluate the influence of factors as the instrument and/or the analyst or over a longer period of time (i.e., typically on different days)(ISO 22412 2017). It is suggested to investigate the trueness of the developed procedure. However, no limit was provided for the relative uncertainty of trueness (ISO 22412 2017) . According to the literature, the limits of the relative uncertainties of intermediate precision and trueness may be set at 5% and 10% for intermediate precision and trueness, respectively.

Fig. 23.3
figure 3

Experimental nested design to investigate the precision of procedure. The factors days, samples, and replicates are studied and the symbols a, b, and n correspond to the number of levels of a nested factor within the factor above ranked

Full size image

Qualified size measurements should be provided to characterize unknown NMs by DLS under quality control conditions. The procedure proposed by Varenne et al. included (1) the control of the absorption spectrum of NMs for ensuring that no absorption band appears at the wavelength of the laser source of the measurement instrument, (2) the evaluation of the optimal concentration of the dispersions of NMs, and (3) the measure of the size of unknown dispersions of NMs at the determined optimal concentration under operational or performance qualification of the instrument (Varenne et al. 2015b, 2016b). It means that the size of CRM whose size and nature is close to the size of the investigated NMs should be measured before and after the evaluation of the size of investigated NMs permitting to evaluate the size of monodispersed NMs under conditions compatible with quality control assessments.

Validated size measurement procedures using DLS proposed by Varenne et al. were suitable to measure the size of a wide range of NMs including polymer NPs, liposomes, and inorganic NPs as silica NPs (Varenne et al. 2015b, 2016b). However, it was found unsuitable to evaluate the size of NPs having a high density such as anastase TiO2 and magnetic NPs whose sizes are in the upper limit of the measurement instrument (Varenne et al. 2019b).

4.2 Evaluation of the Particle Size Distribution

Evaluating the PSD of a dispersion of NMs is a difficult issue. Several size measurement methods present major inherent limitations that hamper reliable determination of the PSD of NM dispersions that have a wide or complex PSD. Besides, there is only one multimodal CRM (ERM-FD102) including silica particles of two sizes, 18.2 and 84 nm certified by EM . However, this CRM is not certified to be used for the determination of PSD. Without an appropriate reference dispersion of NMs, the performance of a method applied to measure PSD cannot be evaluated. No official procedure has been proposed to characterize the PSD of NMs. The scientific community recommended to apply two methods at least based on two different physical principles. One of the methods should be based on a direct size measurement method including image analysis of particles obtained from AFM , SEM , or TEM or it should include a separative size stage combined with batch size measurement method as detector (Varenne et al. 2016a; Caputo et al. 2019; Rice et al. 2013).

It is noteworthy that the DLS method needs to be used with caution while applied to characterize size and PSD of unknown NMs although this technique is widely used in routine. The intensity of the scattered light is proportional to the power six of the radius of NMs. Thus, the intensity of the scattered light due to the large NMs can cover the signal produced by the smaller NMs of the dispersion. Important bias was reported with this method when it is applied for the determination of the size and PSD of NMs having a wide or complex size distribution although it is reliable while applied to the characterization of NMs having a narrow size distribution (Varenne et al. 2016a; Marucco et al. 2019).Direct size measurement methods include EM and AFM (Varenne et al. 2016a; Rice et al. 2013; Song et al. 2009). The size of the NMs is measured directly from images obtained for the NMs. The preparation of samples for observations by EM and AFM consists in the spreading of the NMs on a sample holding (Ghomrasni et al. 2020; Delvallée et al. 2015). This preparation is critical for the quality of the subsequent image analysis process used to determine PSD and may require that a specific procedure may be developed for each NM (Varenne et al. 2020). NMs of the dispersion must be randomly distributed on the surface of the sample holder (carbon grid or mica substrate). It is also preferable that NMs will be well individualized to avoid distortions due to the proximity of neighbor NMs (Fig 23.4a) and to facilitate image processing measurements. It may be difficult to obtain a random deposition of NMs on sample holder from a dispersion of NMs having a high PSD as a segregation according to the NM size may occur as illustrated in the Fig.23.4b. For this reason, it is recommended to evaluate the PSD performing orthogonal measurements with different methods (Varenne et al. 2016a; Sokolova et al. 2011; Cascio et al. 2014; Anderson et al. 2013; Caputo et al. 2019; Ingebrigtsen and Brandl 2002).

Fig. 23.4
figure 4

Electron micrograph of unstained poly(isobutyl cyanoacrylate) NPs deposited on a formvar-carbon coated cupper grid for EM . (a) Projected image of single particles appeared circular suggesting that the particles were spherical. In contrast, particles included in agglomerates appeared distorted due to the close contact with their neighbors. Scale bar: 100 nm. (b) Segregation according to particle size occurred during sample preparation of a highly polydisperse dispersion of the NPs. Scale bare: 2 μm. Evaluation of shape, size, and PSD by EM requires that NPs will be well individualized on the sample holder and randomly distributed over the surface of the sample holder

Full size image

To evaluate PSD from direct methods, size measurements must be performed on a sufficiently large number of NMs. The debate around the number of NMs that should be considered remains open. The ISO standard suggests that the size of one thousand individual NMs should be measured that seems not always possible to achieve due to sample preparation constrains (ISO 13322-1 2004). A much lower number of NMs was considered in different works. Song et al. studied the PSD of a dispersion of synthetic gold NPs consisting in one population of size with a polydisperse distribution and showed that the PSD provided by counting a few hundred NPs was similar to the one produced by the analysis of one thousand NPs (Song et al. 2009). Varenne et al. investigated the PSD of a multimodal dispersion of polymer NPs from the “real-life” obtaining similar PSD from three independent evaluations performed by measuring samples including around three hundred NPs (Varenne et al. 2020). A number of at least five hundred NPs was considered in an interlaboratory comparison of the evaluation of the PSD of NPs performed by TEM indicating a good performance of the method considering this number of NPs (Rice et al. 2013). Rice et al. have found that the best model to use to interpret raw data evaluating the PSD was the lognormal reference model as it provided with the lower relative standard errors (RSEs) compared with other size distribution reference models tested in their work while determining the PSD of their NM (Rice et al. 2013).

4.3 Evaluation of Zeta Potential Using Electrophoresis Light Scattering

Reliable evaluation of ZP of NMs by electrophoretic light scattering (ELS) also requires the validation of measurement procedure s. An ISO standard gives guidelines for good practices in the evaluation of ZP (ISO/TR 19997 2018). Another ISO standard indicates the thresholds for the relative standard uncertainties of repeatability, intermediate precision , and trueness that should be used for the validation of procedures to evaluate ZP (ISO 13099-2 2012).

A similar strategy than that applied to validate the procedure of size measurements may be applied (Varenne et al. 2015a, c, 2019a, b). In short, as for size measurements performed by DLS, the preparation of samples to evaluate ZP by ELS is a key step (Varenne et al. 2015a, c, 2019a, b). The presence of dust in samples can be avoided preparing dilutions with fresh filter dispersants with 0.22 μm filter just before use. All flasks with caps devoted to the preparation of dispersant and samples are needed to be pre-cleaned with filtered ultrapure water and stored in a dust-free environment. Selection of high-quality measurement cell is needed as optical defects including scratches and/or apparent impurities in the polycarbonate faces may interfere with optical measurements. Beside cell cleanliness appearance, electrodes should be homogenous and well attached on both the inside and outside of the cell measurement to insure a homogeneous electric field. The cells including caps should be rinsed with appropriate filtered solvent and stored in a dust-free environment before using. The cells should be used only once to prevent cross-contamination. The temperature of the sample is a critical parameter. Large differences between the temperature of the sample and the temperature of measurement may generate artefacts during measurements due to the degassing of the samples.

Optimal concentration of the dispersions of NMs to evaluate ZP should be evaluated using methods based on the equilibrium dilution procedure mentioned in the ISO standard (ISO 13099-1 2012). This procedure consists in maintaining the composition and the concentration of dispersant identical between diluted samples.

The quality of data may be appreciated by means of defined quality criteria achieved during measurement and on the raw data (Varenne et al. 2015a, c). For example, the phase plot showing phase difference between the measured frequency and the reference frequency as a function of time and the count rate curve giving the number of photons detected by the photomultiplier associated to each run can be inspected during measurement. The final phase plot, the frequency plot corresponding to the Fourier Transform analysis of the slow field reversal part of the analysis used to evaluate ZP distribution and the mean count rate, can be controlled on the raw data.

Experimental measurement procedures established to evaluate the ZP of an NM must be validated using reference NMs. Only two were developed so far. One CRM is available with assigned SI-traceable values of positive electrophoretic mobility (NIST Standard Reference Material® 1980). It is noteworthy that this CRM tends to adsorb on the intern surface of measurement cells made of polycarbonate (Varenne et al. 2015a). For this reason, measurement cells should be preconditioned with the dilute dispersions of NMs before introducing fresh samples and carrying out the analysis as explained in the notice of use. To validate procedures for NMs with a negative ZP, it necessary to use one negative ZP RM classified as a transfer standard. This type of standard has been referenced to an accepted standard by the scientific community as there is no CRM with acceptable uncertainty of the certified value of ZP (Braun et al. 2011b).

The procedures should be validated by investigating robustness, precision , that is, repeatability and intermediate precision, and trueness to determine the expanded uncertainties of the procedures. The same experimental design than the one presented in Fig. 23.2 may be used to investigate the precision of the developed procedures. The repeatability can be determined with successive evaluation of ZP of the RM on the same day while the intermediate precision can be assessed by carrying out the evaluation of ZP of the RM for various days. The ISO standard gives thresholds for the relative standard uncertainties of repeatability, intermediate precision, and trueness (10%, 15%, and 10%, respectively) (ISO 13099-2 2012).

Qualified evaluation of ZP can be performed following the same procedure than the one described for the measurement of size of NMs. According to the mode used to perform the evaluation of ZP, the proposed procedures by Varenne et al. can be applied to the characterization of NMs including polymer NPs and liposomes, but were not appropriate to evaluate the ZP of dense NPs such as titanium dioxide NPs (Varenne et al. 2015a, 2019a, b). In any case, the evaluation of ZP of an NM is not trivial as many parameters can influence the final results (Skoglund et al. 2017). A series of advices on how to interpret and report measurements of ZP was proposed based on the evaluation of the ZP of metal NPs dispersed in complex media of relevance for studies on nanotoxicology and environmental interactions (Skoglund et al. 2017).

4.4 Transfer

Once validated in one laboratory, it has to be demonstrated that the validated procedure can be applied in other laboratories with the same performances. A transfer of the procedure is needed to prove that the results of measurements are similar in all laboratories. Such transfer was achieved for very few procedures applied to the characterization of NMs (Weigel et al. 2017; Langevin et al. 2018a, b; Varenne et al. 2017; Franks et al. 2019). For instance, procedures to characterize size and ZP of NMs by DLS and ELS, respectively, were transferred from one sending laboratory to other laboratories (Varenne et al. 2017). Two situations were considered. In the first case, the sending and receiving laboratories were equipped with the same measurement instrument (same wavelength of the laser source). A comparative test performed on the same batch of CRM or RM was proposed to show that performances of the receiving laboratories were similar to those of the sending laboratory taking into account handling precautions, crucial factor highlighted by the validation carried out by the sending laboratory and measurement quality criteria. In the second case, the sending and receiving laboratories were not equipped with the same instrument (different wavelength of the laser source). It was then suggested to perform a partial validation to prove the ability of receiving laboratories to perform the procedures. This partial validation was based on the study of the precision , that is, repeatability and intermediate precision, and the trueness to assess the expanded uncertainties of the procedure. To achieve the transfer of a procedure, it is important that all partners carry on measurements of the same batch of CRM or RM . Results of measurements obtained by the different laboratories are compared using statistical analytical methods. A development of appropriate methods was proposed in the work of Varenne et al. based on the β-expectation tolerance interval method and ANOVA (Varenne et al. 2017) .

5 Conclusion

Main issues found for the characterization of NMs were considered in the present chapter. It discussed the validation of analytical procedures based on metrology approaches to be applied to assess the quality analysis of NMs. The reflexion associated basis in metrology and their application to the method of characterization of the main physicochemical parameters that are used to define NMs. This analysis pointed out the urgent need to standardize, validate, and transfer analytical procedures applied to characterize NMs. This is paramount to ensure the reliability of results obtained from the quality assessment of NMs which, in turn, is needed to ensure their safety providing proof of the repeatability and efficiency of industrial processes producing NM-based products. Today, physicochemical characterization of NMs associated with metrology remains a challenge for future development in all application fields. Quality assessment of NMs is still in its infant age. Efforts are on the way to provide with more official guidelines to perform validation and transfer of measurement procedure s and develop appropriate RM including CRM . Besides, several validated measurement procedures and results from interlaboratory measurement comparisons were published in the literature that can now serve as basis to go further setting up quality control procedures for NMs.