Electrically assisted capillary liquid chromatography using a silica monolithic column

A silica monolithic capillary column was linked to an open capillary of the same internal diameter via a Teflon sleeve to form a duplex column to investigate the combination of chromatography and electrophoresis in the mode of electrically assisted capillary liquid chromatography (eCLC). Using a commercial CE instrument with an 8.5 cm long, 100 μm i.d. reversed phase silica monolithic section and a window 1.5 cm beyond the end of this in a 21.5 cm open section, a minimum plate height of 9 μm was obtained in capillary liquid chromatography (CLC) mode at a low driving pressure of 50 psi. In eCLC mode, high speed and high resolution separations of acidic and basic compounds were achieved with selectivity tuning based on the flexible combination of pressure (0–100 psi) and voltage. Taking advantage of the excellent permeability of silica monolithic columns, use of a step flow gradient enabled elution of compounds with different charge state.     

Simple capillary flow porometer for characterization of capillary columns containing packed and monolithic beds

A simple capillary flow porometer (CFP) was assembled for through-pore structure characterization of monolithic capillary liquid chromatography columns in their original chromatographic forms. Determination of differential pressures and flow rates through dry and wet short capillary segments provided necessary information to determine the mean diameters and size distributions of the through-pores. The mean through-pore diameters of three capillary columns packed with 3, 5, and 7 μm spherical silica particles were determined to be 0.5, 1.0 and 1.4 μm, with distributions ranging from 0.1 to 0.7, 0.3 to 1.1 and 0.4 to 2.6 μm, respectively. Similarly, the mean through-pore diameters and size distributions of silica monoliths fabricated via phase separation by polymerization of tetramethoxysilane (TMOS) in the presence of poly(ethylene glycol) (PEG) verified that a greater number of through-pores with small diameters were prepared in columns with higher PEG content in the prepolymer mixture. The CFP system was also used to study the effects of column inner diameter and length on through-pore properties of polymeric monolithic columns. Typical monoliths based on butyl methacrylate (BMA) and poly(ethylene glycol) diacrylate (PEGDA) in capillary columns with different inner diameters (i.e., 50–250 μm) and lengths (i.e., 1.5–3.0 cm) were characterized. The results indicate that varying the inner diameter and/or the length of the column had little effect on the through-pore properties. Therefore, the through-pores are highly interconnected and their determination by CFP is independent of capillary length.     

Rapid LC-MS detection of cyanobacterial hepatotoxins microcystins and nodularins—Comparison of columns

Eight reversed-phase columns intended for rapid HPLC were assessed for the separation of thirteen microcystins and nodularins, cyclic peptidic hepatotoxins. The instrumentation consisted of an Agilent Technologies 1200 Rapid Resolution high performance liquid chromatography system coupled to a mass spectrometer, Bruker Daltonics Ultra Performance High Capacity Ion Trap MS (HCT Ultra) with electrospray ionisation (RRLC-ESI-IT-MS). The columns tested were 2-2.1 mm × 50 mm in diameter and length, and contained small particles (1.8-2.7 μm), or monolithic silica supports for fast performance. The shortest total run time achieved was 3 min 15 s including equilibration and injection. Critical microcystin pairs were still resolved. Several columns showed excellent performance.     

Radial heterogeneity of some analytical columns used in high-performance liquid chromatography

An on-column electrochemical microdetector was used to determine accurately the radial distribution of the mobile phase velocity and of the column efficiency at the exit of three common analytical columns, namely a 100 mm × 4.6 mm C18 bonded silica-based monolithic column, a 150 mm × 4.6 mm column packed with 2.7 μm porous shell particles of C18 bonded silica (HALO), and a 150 mm × 4.6 mm column packed with 3 μm fully porous C18 bonded silica particles (LUNA). The results obtained demonstrate that all three columns are not radially homogeneous. In all three cases, the efficiency was found to be lower in the wall region of the column than in its core region (the central core with a radius of 1/3 the column inner radius). The decrease in local efficiency from the core to the wall regions was lower in the case of the monolith (ca. 25%) than in that of the two particle-packed columns (ca. 35–50%). The mobile phase velocity was found to be ca. 1.5% higher in the wall than in the core region of the monolithic column while, in contrast, it was ca. 2.5–4.0% lower in the wall region for the two particle-packed columns.     

The performance of hybrid monolithic silica capillary columns prepared by changing feed ratios of tetramethoxysilane and methyltrimethoxysilane

The effect of a feed ratio of methyltrimethoxysilane (MTMS) to tetramethoxysilane (TMOS) was studied to improve the performance of a hybrid monolithic silica capillary column with 100-μm i.d. in HPLC in a range MTMS/TMOS (v/v) = 10/90–25/75. The domain size was also varied by adjusting the amount of PEG to control permeability (K = 2.8 × 10⁻¹⁴–6.9 × 10⁻¹⁴ m²). Evaluation of the performance for those capillary columns following octadecylsilylation proved an increase in retention factor (k) and a decrease in steric selectivity α(triphenylene/ortho-terphenyl) with the increase in MTMS content in the feed. The effect of the feed ratio was also observed in porosity and hydrophobic property of the C18 stationary phase from the results of size exclusion chromatography (SEC) and reversed phase characterization. The monolithic silica capillary columns prepared under new preparation conditions were able to produce a plate height of 4.6–6.0 μm for hexylbenzene in a mobile phase acetonitrile/water = 80/20 at a linear velocity of 2 mm/s. Consequently, it was possible to prepare hybrid monolithic silica capillary columns with higher performance than those reported previously while maintaining the retention factors in a similar range by reducing the MTMS/TMOS ratio and increasing the total silane concentration in feed.     

Parameters affecting the separation of intact proteins in gradient-elution reversed-phase chromatography using poly(styrene-co-divinylbenzene) monolithic capillary columns

An experimental study was performed to investigate the effects of column parameters and gradient conditions on the separation of intact proteins using styrene-based monolithic columns. The effect of flow rate on peak width was investigated at constant gradient steepness by normalizing the gradient time for the column hold-up time. When operating the column at a temperature of 60 °C a small C-term effect was observed in a flow rate range of 1–4 μL/min. However, the C-term effect on peak width is not as strong as the decrease in peak width due to increasing flow rate. The peak capacity increased according to the square root of the column length. Decreasing the macropore size of the polymer monolith while maintaining the column length constant, resulted in an increase in peak capacity. A trade-off between peak capacity and total analysis time was made for 50, 100, and 250 mm long monolithic columns and a microparticulate column packed with 5 μm porous silica particles while operating at a flow rate of 2 μL/min. The peak capacity per unit time of the 50 mm long monolithic column with small pore size was superior when the total analysis time is below 120 min, yielding a maximum peak capacity of 380. For more demanding separations the 250 mm long monolith provided the highest peak capacity in the shortest possible time frame.     

Practical aspects of fast HPLC separations for pharmaceutical process development using monolithic columns

A large number of samples can be generated during pharmaceutical process development. Fast separation for these samples is usually challenging due to the complexity of sample matrix, which requires high efficiency as well as high speed. Monolithic columns (E. Merck, Germany) were investigated as a possible tool for reducing separation time in reversed-phase HPLC without significantly sacrificing efficiency or resolution. Both van Deemter plots and separations of alkyl benzenes and in-process samples showed that monolithic columns were suitable for fast separations without significantly compromising resolution. Practical parameters including the pressure drop, retention factor, selectivity, and tailing factor of monolithic columns (Chromolith type) were compared to those of conventional YMC 150 mm × 4.6 mm (3-μm particles) and 250 mm × 4.6 mm (5-μm particles) packed columns. The batch-to-batch reproducibility of the 100 mm × 4.6 mm Chromolith columns from five randomly ordered batches was also compared to the 250 mm × 4.6 mm YMC particle-packed columns. Fast and efficient separations of complicated process samples including crude drug substances, reaction mixtures, and crystallized mother liquors were demonstrated for both monolithic columns and conventional packed columns. The analysis times were decreased by three to seven times on the coupled monolithic columns, while maintaining the comparable resolution to typical 5-μm particle-packed 250 mm × 4.6 mm columns.     

Flow-through pore characteristics of monolithic silicas and their impact on column performance in high-performance liquid chromatography

In order to elucidate the role of the flow-through characteristics with regard to the column performance in high-performance liquid chromatography (HPLC) native and n-octadecyl bonded monolithic silica rods and columns, respectively of 100 mm length and 4.6 mm ID with mesopores in the range between 10 and 25 nm and macropores in the range between 0.7 and 6.0 μm were examined by mercury intrusion/extrusion, scanning electron microscopy, image analysis and permeability. The obtained data of the flow-through pore sizes and porosity values as well as surface-to-volume ratio of the stationary phase skeleton enabled to predict their influence to the chromatographic separation efficiency. Our data demonstrate that mercury porosimetry is a reliable technique to obtain all the characteristic parameters of the flow-through pores of silica monoliths. An important result of our examination was that the surface-to-volume ratio of monolithic silica skeletons had more significant impact to the separation process, rather than the average flow-through pore sizes. We could also show the essential differences between the particulate and monolithic stationary phases based on theoretical computation. The results, obtained from other characterization methods also indicated the structural complexity of monolithic silica samples. Permeability of columns is a generally applicable parameter to characterize all chromatographic phases no matter the chemistry or format. The correlation coefficient obtained for mercury intrusion and permeability of water was 0.998, though our investigation revealed that the surface modification is more likely influencing the obtained results. Further, the assumption of the cylindrical morphology of flow-through pores is not relevant to the investigated monolithic silica columns. These results on the morphology of the flow-through pores and of the skeletons were confirmed by the image analysis as well. Our main finding is that the flow-through pore sizes are not relevant for the estimation of the chromatographic separation efficiency of monolithic silica columns.     

High-performance computing of flow and transport in physically reconstructed silica monoliths

This work presents an approach towards resolving hydrodynamic flow in real porous media by carrying out direct numerical simulations in the reconstructed macroporous (flow-through) domain of a silica monolith. The macroporous domain of a 60 μm × 60 μm × 12 μm segment of a 100 μm i.d. capillary silica monolith was reconstructed by confocal laser scanning microscopy. A 60 μm × 12 μm × 12 μm segment of the reconstructed domain was then used as the 3D matrix for simulation of fluid flow by the lattice-Boltzmann method on a high-performance computing platform. Excellent agreement is observed between the experimental and simulated Darcy permeabilities without any assumptions or further adjustments on the monolith morphology. The flow velocity field is analyzed in detail, including longitudinal and transverse velocity distributions, the occurrence of negative longitudinal velocities, as well as the beginning transition to the viscous-inertial flow regime. The presented methodology promises great potential for resolving the key relationships between morphology and band broadening in monolithic columns for HPLC applications.     

Capillary electrochromatography with contactless conductivity detection for the determination of some inorganic and organic cations using monolithic octadecylsilica columns

A fast separation of alkali and alkaline earth metal cations and ammonium was carried out by capillary electrochromatography on monolithic octadecylsilica columns of 15 cm length and 100 μm inner diameter using water/methanol mixtures containing acetic acid as mobile phase. On-column contactless conductivity detection was used for quantification of these non-UV-absorbing species. The method was also extended successfully to the determination of small amines as well as of amino acids, and the separation selectivity was optimized by varying the composition of the mobile phase. Detection limits of about 1 μM were possible for the inorganic cations as well as for the small amines, while the amino acids could be quantified down to about 10 μM. The separation of 12 amino acids was achieved in the relatively short time of 10 min.     

Improvement of separation efficiencies of anion-exchange chromatography using monolithic silica capillary columns modified with polyacrylates and polymethacrylates containing tertiary amino or quaternary ammonium groups

Anion-exchange (AEX) columns were prepared by on-column polymerization of acrylates and methacrylates containing tertiary amino or quaternary ammonium groups on monolithic silica in a fused silica capillary modified with anchor groups. The columns provided a plate height (H) of less than 10 μm at optimum linear velocity (u) with keeping their high permeability (K = 9–12 × 10⁻¹⁴ m²). Among seven kinds of AEX columns, a monolithic silica column modified with poly(2-hydroxy-3-(4-methylpiperazin-1-yl)propyl methacrylates) (HMPMA) showed larger retentions and better selectivities for nucleotides and inorganic anions than the others. The HMPMA column of 410 mm length produced 42 000–55 000 theoretical plates (N) at a linear velocity of 0.97 mm/s with a backpressure of 3.8 MPa. The same column could be employed for a fast separation of inorganic anions in 1.8 min at a linear velocity of 5.3 mm/s with a backpressure of 20 MPa. In terms of van Deemter plot and separation impedance, the HMPMA column showed higher performance than a conventional particle-packed AEX column. The HMPMA column showed good recovery of a protein, trypsin inhibitor, and it was applied to the separation of proteins and tryptic digest of bovine serum albumin (BSA) in a gradient elution, to provide better separation compared to a conventional particle-packed AEX column.     

Intact-protein trapping columns for proteomic analysis in capillary high-performance liquid chromatography

A new type of monolithic trapping columns with high mechanical strength was prepared by thin-layer sol–gel coating method and applied to trapping intact proteins for on-line capillary liquid chromatography. Monolithic trapping columns were fabricated by entrapping C8 reversed-phase particles into the capillary columns through a sol–gel network, which was formed by hydrolysis and polycondensation of methyltriethoxysilane. Hundreds times of trapping/untrapping for intact proteins were carried out. The trapping columns showed long-term stability up to 300 bar. Recovery, loading capacity and reproducibility of trapping columns were evaluated using four proteins. The recovery of four protein mixtures for the C8 monolithic trapping columns was 99.3% on average. The loading capacity of 5 mm × 320 μm i.d. C8 trapping columns for the protein mixtures was 30 μg. Day-to-day relative standard deviation (RSD) values for recoveries of protein mixtures on the same C8 trapping column ranged from 2.34 to 5.87%, column-to-column RSD values were from 3.01 to 6.81%. The C8 trapping columns were used to trap normal mouse liver intact proteins in a capillary liquid chromatography system. Results demonstrated high efficiency of the monolithic trapping columns for trapping intact proteins for proteomic analysis in on-line capillary liquid chromatography system.     

Polymethacrylate monolithic and hybrid particle-monolithic columns for reversed-phase and hydrophilic interaction capillary liquid chromatography

We prepared hybrid particle-monolithic polymethacrylate columns for micro-HPLC by in situ polymerization in fused silica capillaries pre-packed with 3–5 μm C₁₈ and aminopropyl silica bonded particles, using polymerization mixtures based on laurylmethacrylate–ethylene dimethacrylate (co)polymers for the reversed-phase (RP) mode and [2-(methacryloyloxy)ethyl]-dimethyl-(3-sulfopropyl) zwitterionic (co)polymers for the hydrophilic interaction (HILIC) mode. The hybrid particle-monolithic columns showed reduced porosity and hold-up volumes, approximately 2–2.5 times lower in comparison to the pure monolithic columns prepared in the whole volume of empty capillaries. The elution volumes of sample compounds are also generally lower in comparison to packed or pure monolithic columns. The efficiency and permeability of the hybrid columns are intermediate in between the properties of the reference pure monolithic and particle-packed columns. The chemistries of the embedded solid particles and of the interparticle monolithic moiety in the hybrid capillary columns contribute to the retention to various degrees, affecting the selectivity of separation. Some hybrid columns provided improved separations of proteins in comparison to the reference particle-packed columns in the reversed-phase mode. Zwitterionic hybrid particle-monolithic columns show dual mode retention HILIC/RP behaviour depending on the composition of the mobile phase and allow separations of polar compounds such as phenolic acids in the HILIC mode at lower concentrations of acetonitrile and, often in shorter analysis time in comparison to particle-packed and full-volume monolithic columns.     

Monolithic silica spin column extraction and simultaneous derivatization of amphetamines and 3,4-methylenedioxyamphetamines in human urine for gas chromatographic-mass spectrometric detection

A simple, sensitive, and specific method with gas chromatography-mass spectrometry was developed for simultaneous extraction and derivatization of amphetamines (APs) and 3,4-methylenedioxyamphetamines (MDAs) in human urine by using a monolithic silica spin column. All the procedures, such as sample loading, washing, and elution were performed by centrifugation. APs and MDAs in urine were adsorbed on the monolithic silica and derivatized with propyl chloroformate in the column. Methamphetamine-d₅ was used as an internal standard. The linear ranges were 0.01-5.0 μg mL⁻¹ for methamphetamine (MA) and 3,4-methylenedioxymethamphetamine (MDMA) and 0.02-5.0 μg mL⁻¹ for amphetamine (AP) and 3,4-methylenedioxyamphetamine (MDA) (coefficient of correlation ≧0.995). The recovery of APs and MDAs in urine was 84-94%, and the relative standard deviation of the intra- and interday reproducibility for urine samples containing 0.1, 1.0, and 4.0 μg mL⁻¹ of APs and MDAs ranged from 1.4% to 13.6%. The lowest detection limit (signal-to-noise ratio ≧ 3) in urine was 5 ng mL⁻¹ for MA and MDMA and 10 ng mL⁻¹ for AP and MDA. The proposed method can be used to perform simultaneous extraction and derivatization on spin columns that have been loaded with a small quantity of solvent by using centrifugation.     

Enhanced capabilities of separation in Sequential Injection Chromatography--fused-core particle column and its comparison with narrow-bore monolithic column

In the Sequential Injection Chromatography (SIC) only monolithic columns for chromatographic separations have been used so far. This article presents the first use of fused-core particle packed column in an attempt to extend of the chromatographic capabilities of the SIC system. A new fused-core particle column (2.7 μm) Ascentis^® Express C18 (Supelco™ Analytical) 30 mm × 4.6 mm brings high separation efficiency within flow rates and pressures comparable to monolithic column Chromolith^® Performance RP-18e 100-3 (Merck^®) 100 mm × 3 mm. Both columns matches the conditions of the commercially produced SIC system - SIChrom™ (8-port high-pressure selection valve and medium-pressure Sapphire™ syringe pump with 4 mL reservoir - maximal work pressure 1000 PSI) (FIAlab^®, USA). The system was tested by the separation of four estrogens with similar structure and an internal standard - ethylparaben. The mobile phase composed of acetonitrile/water (40/60 (v/v)) was pumped isocratic at flow rate 0.48 mL min⁻¹. Spectrophotometric detection was performed at wavelength of 225 nm and injected volume of sample solutions was 10 μL. The chromatographic characteristics of both columns were compared. Obtained results and conclusions have shown that both fused-core particle column and longer narrow shaped monolithic column bring benefits into the SIC method.     

Practical assessment of frictional heating effects and thermostat design on the performance of conventional (3 μm and 5 μm) columns in reversed-phase high-performance liquid chromatography

A practical investigation of frictional heating effects in conventional C18 columns was undertaken, to investigate whether problems found for sub-2 μm columns were also present for those of particle size 3 μm and 5 μm and different internal diameter. The influence of a water bath, a still air heater, and a forced air heater on performance was investigated. Heating effects were substantial, with a decrease in k of almost 15% for toluene over the flow rate range ∼0.4–2.3 mL/min with a 15 cm × 0.46 cm ID column packed with 3 μm particles. Heating effects on retention increased with increasing solute k, with increase in the column ID, with decrease in the column particle size, and with decrease in the set column oven temperature. While the water bath minimised axial temperature gradients and thus its effect on k, radial temperature gradients were potentially serious with this system, especially at high mobile phase velocity, even with columns containing 5 μm particles. In contrast to the effects of axial temperature gradients in 4.6 mm columns, very little difference in Van Deemter plots was noted between the three different thermostats with 2 mm ID columns, even when 3 μm particles were used. However, the efficiency of 2 mm columns for peaks of low or moderate k (k < 4) can be compromised by the extra dead volume introduced by the heating systems, even with conventional HPLC systems with otherwise minimised extra column volume.     

Ring-opening metathesis polymerization for the preparation of norbornene-based weak cation-exchange monolithic capillary columns

Functionalized monolithic columns were prepared via ring-opening metathesis polymerization (ROMP) within silanized fused silica capillaries with an internal diameter of 200 μm by in situ grafting. This procedure is conducted in two steps, the first of which is the formation of the basic monolithic structure by polymerization of norborn-2-ene (NBE) and 1,4,4a,5,8,8a-hexahydro-1,4,5,8-exo,endo-dimethanonaphthalene (DMN-H6) in a porogenic system (toluene and 2-propanol) using RuCl₂(PCy₃)₂(CHPh) as ROMP initiator. In the second step the still active initiator sites located on the surface of the structure-forming microglobules were used as receptor groups for the attachment (“grafting”) of functional groups onto the monolithic backbone by flushing the monolith with 7-oxanorborn-2-ene-5,6-carboxylic anhydride (ONDCA). Functionalization conditions were first defined that did not damage the backbone of low polymer content (20%) monoliths allowing high-throughput chromatographic separations. Variation of the functionalization conditions was then shown to provide a means of controlling the degree of functionalization and resulting ion-exchange capacity. The maximum level of in situ ONDCA grafting was obtained by a 3 h polymerization in toluene at 40 °C. The weak cation-exchange monoliths obtained provided good separation of a standard peptide mixture comprising four synthetic peptides designed specifically for the evaluation of cation-exchange columns. An equivalent separation was also achieved using the lowest capacity column studied, indicative of a high degree of robustness of the functionalization procedure. As well as demonstrably bearing ionic functional groups enabling analyte separation in the cation-exchange mode, the columns exhibited additional hydrophobic characteristics which influenced the separation process. The functionalized monoliths thus represent useful tools for mixed-mode separations.     

Silica-based monolithic capillary columns—Effect of preparation temperature on separation efficiency

The temperature effects during the sol–gel process and ageing of the silica-based monolith on the structure and separation efficiency of the capillary columns (100 μm i.d., 150 mm) for HPLC separations were studied. The tested columns were synthesized from a mixture of tetramethoxysilane, polyethylene glycol and urea under the acidic conditions. The temperature was varied from 40 °C to 44 °C and formation of bypass channels between the silica mold and the capillary wall was examined. The temperature of 43 °C was estimated as optimal for preparation of efficient silica capillary columns which were subsequently modified by octadecyldimethyl-N,N-diethylaminosilane or covered by poly(octadecyl methacrylate) and tested using standard mixture of alkylbenzenes under the isocratic conditions.     

Preparation and evaluation of open tubular C18-silica monolithic microcartridges for preconcentration of peptides by on-line solid phase extraction capillary electrophoresis

In this study, C18-silica monoliths were synthesized as a porous layer in open tubular capillary columns, to be cut later into microcartridges for the analysis of neuropeptides by on-line solid-phase extraction capillary electrophoresis with UV and MS detection (SPE-CE-UV and SPE-CE-MS). First, several types of C18-silica monolithic (MtC18) microcartridges were used to analyse standard solutions of five neuropeptides (i.e. dynorphin A (1–7), substance P (7–11), endomorphin 1, methionine enkephalin and [Ala]-methionine enkephalin). The MtC18 sorbents were especially selective against endomorphin 1 and substance P (7–11)). The best results in terms of sensitivity and inter-microcartridge reproducibility were achieved with the microcartridges obtained from a 10-cm open tubular capillary column with a thin monolithic coating with large through-pores (1–5 μm). Run-to-run repeatability, microcartridge durability, linearity ranges and LODs were studied by MtC18-SPE-CE-MS. As expected due to their greater selectivity, the best LOD enhancement was obtained for End1 and SP (7–11) (50 times with regard to CE-MS). Finally, the suitability of the methodology for analysing biological fluids was tested with plasma samples spiked with End1 and SP (7–11). Results obtained were promising because both neuropeptides could be detected at 0.05 μg mL⁻¹, which was almost the same concentration level as for the standard solutions (0.01 μg mL⁻¹).     

Fast gradient screening of pharmaceuticals with 5 cm long, narrow bore reversed-phase columns packed with sub-3 μm core-shell and sub-2 μm totally porous particles

The performance of 5 cm long narrow-bore columns packed with 2.6-2.7 μm core-shell particles and a column packed with 1.7 μm totally porous particles was compared in very fast gradient separations of polar neutral active pharmaceutical compounds. Peak capacities as a function of flow-rate and gradient time were measured. Peak capacities around 160-170 could be achieved within 25 min with these 5 cm long columns. The highest peak capacity was obtained with the Kinetex column however it was found that as the flow-rate increases, the peak capacity of the new Poroshell-120 column is getting closer to that obtained with the Kinetex column. Considering the column permeability, peak capacity per unit time and per unit pressure was also calculated. In this comparison the advantage of sub-3 μm core-shell particles is more significant compared to sub-2 μm totally porous particles. Moreover it was found that the very similar sized (d_p = 2.7 μm) and structured (ρ = 0.63) new Poroshell-120 and the earlier introduced Ascentis Express particles showed different efficiency. Results obtained showed that the 5 cm long narrow bore columns packed with sub-3 μm core-shell particles offer the chance of very fast and efficient gradient separations, thus these columns can be applied for fast screening measurements of routine pharmaceutical analysis such as cleaning validation.