Droplet-based microfluidics detector for bioaerosol detection

Detection of bioaerosols is important in fields ranging from environmental health monitoring to biosurveillance, and current detector weaknesses have motivated the development of new technologies. In this work, a detector was built, which applies the principles of droplet microfluidics to bioaerosol detection. Droplet microfluidics is a subfield of microfluidics based on the creation of monodisperse microdroplets with compartmentalized reagents and supports enhanced assays and fluidic manipulations. The bioaerosol detector operates by aerodynamically focusing aerosols directly into these droplets to harness the benefits of the microreactor environment. A breadboard detector system, which consisted of an aerodynamic focusing lens, aerosol-focusing capillary, microfluidic droplet chip, and optical microscope, was constructed. Computational fluid dynamic simulations and Lagrangian particle tracking modeling were conducted to identify the optimal conditions for focusing. Preliminary experiments, where aerosols were deposited onto a solid substrate, demonstrated sub 200-µm spot diameters for aerodynamic diameters of 2–5 µm. Test aerosols were then generated, and collected into the microfluidic liquid interface on the chip as verified by microscopy. Recovery efficiency of the aerosols was dependent on aerosol size and ranged from about 27% to nearly 100%. Finally, to prove bioaerosol collection and detection, a droplet propidium iodide (PI) assay was performed: the system distinguished between E. coli and non-biological aerosols within 20 s. Overall, this work established the technique of direct collection of bioaerosols into a convenient droplet microfluidic platform for detection.

Copyright © 2017 The Johns Hopkins University Applied Physics Laboratory     

Design and evaluation of an aerodynamic focusing micro-well aerosol collector

Aerosol sampling and identification is vital for the assessment and control of particulate matter pollution, airborne pathogens, allergens, and toxins and their effect on air quality, human health, and climate change. In situ analysis of chemical and biological airborne components of aerosols on a conventional filter is challenging due to dilute samples in a large collection region. We present the design and evaluation of a micro-well (µ-well) aerosol collector for the assessment of airborne particulate matter (PM) in the 0.5–3 µm size range. The design minimizes particle collection areas allowing for in situ optical analysis and provides an increased limit of detection for liquid-based assays due to the high concentrations of analytes in the elution/analysis volume. The design of the collector is guided by computational fluid dynamics (CFD) modeling; it combines an aerodynamic concentrator inlet that focuses the aspirated aerosol into a narrow beam and a µ-well collector that limits the particle collection area to the µ-well volume. The optimization of the collector geometry and the operational conditions result in high concentrations of collected PM in the submillimeter region inside the µ-well. Collection efficiency experiments are performed in the aerosol chamber using fluorescent polystyrene microspheres to determine the performance of the collector as a function of particle size and sampling flow rate. The collector has the maximum collection efficiency of about 75% for 1 µm particles for the flow rate of 1 slpm. Particles bigger than 1 µm have lower collection efficiencies because of particle bounce and particle loss in the aerodynamic focusing inlet. Collected samples can be eluted from the device using standard pipettes, with an elution volume of 10–20 µL. The transparent collection substrate and the distinct collection region, independent of particle size, allows for in situ optical analysis of the collected PM.

© 2017 American Association for Aerosol Research     

Assessment of indoor bioaerosols using a lab-made virtual impactor

To assess indoor bioaerosols, a virtual impactor having 1 µm cutoff diameter was designed, fabricated, and evaluated with computational fluid dynamics simulation and also with laboratory test using polystyrene latex particles. Two other cutoff diameters of 635 nm and 1.5 µm were obtained by changing the inlet flow rate and the ratio of minor channel-to-inlet flow rates. In field test, the virtual impactor was operated with varying cutoff diameter and field-emission scanning electron microscope (FE-SEM) analysis was performed for each cutoff diameter to observe morphologies of indoor aerosol particles sampled at the major and minor outlet channels. Particles were sampled at both outlet channels using the SKC Button Aerosol sampler and subsequently cultured. By colony counting, it was found that 56% of cultured fungal particles and 63% of cultured bacterial particles had aerodynamic sizes smaller than 1 µm. MALDI-TOF analysis and visual inspection of culture samples were used to identify indoor bacterial and fungal species, respectively. Nearly same species of bacteria and fungi were detected both in the major and minor flow channels.

© 2017 American Association for Aerosol Research     

Fluidic and thermal properties of heliox enable the efficient generation and delivery of high concentrations of solid-phase,fine particle aerosols from viscous liquids

In patients with impaired respiratory function, heliox (80% helium, 20% oxygen) has been shown to increase the peripheral deposition of aerosols. It was hypothesized that using SUPRAER-CH, aerosols generated from viscous solutions/suspensions (4–40?cP) can be delivered as solid-phase aerosols with smaller aerodynamic diameters, at higher output efficiencies and increased pay-loads using heliox than with air using SUPRAER-CA. Aerosols were generated from solutions/suspensions of 100?mg/ml bovine serum albumin (BSA), 100?mg/ml bovine gamma globulin, 100?mg/ml polyvinylpirrolidone (PVP), 103?mg/ml and 35?mg/ml surfactant, and delivered at 44?l/min using SUPRAER-CH with heliox and SUPRAER-CA with air. Using SUPRAER-CH with heliox, aerosols of 1.1?µm to 3?µm the mass median aerodynamic diameter (MMAD) were generated from the above agents at output efficiencies between 70% and 88%. Using SUPRAER-CA with air, MMAD of particles delivered from these agents ranged between 2.5?µm and 3.6?µm with output efficiencies between 45% and 65%. Using heliox together with a liquid flow rate of 3?ml/min, the 8?kDa PVP, BSA, bovine gamma globulin, and surfactant aerosols were delivered at 4.3?mg/s, 3.8?mg/s, 3.5?mg/s, and 3.2?mg/s, respectively, with output efficiencies greater than 70%. Up to 2.2?g were collected at the output in 10?min (i.e., 10% 8?kDa PVP). These data indicate the superior utility of heliox to generate fine particle, solid-phase aerosols of proteins, antibodies, and surfactant suitable for delivery to the peripheral lung at clinically relevant doses. The high delivery rates could enable short treatment times.

Copyright © 2019 American Association for Aerosol Research     

Evaluation of the Alphasense optical particle counter (OPC-N2) and the Grimm portable aerosol spectrometer (PAS-1.108)

We compared the performance of a low-cost (～$500), compact optical particle counter (OPC, OPC-N2, Alphasense) to another OPC (PAS-1.108, Grimm Technologies) and reference instruments. We measured the detection efficiency of the OPCs by size from 0.5 to 5 µm for monodispersed, polystyrene latex (PSL) spheres. We then compared number and mass concentrations measured with the OPCs to those measured with reference instruments for three aerosols: salt, welding fume, and Arizona road dust. The OPC-N2 detection efficiency was similar to the PAS-1.108 for particles larger than 0.8 µm (minimum of 79% at 1 µm and maximum of 101% at 3 µm). For 0.5-µm particles, the detection efficiency of the OPC-N2 was underestimated at 78%, whereas PAS-1.108 overestimated concentrations by 183%. The mass concentrations from the OPCs were linear (r ≥ 0.97) with those from the reference instruments for all aerosols, although the slope and intercept were different. The mass concentrations were overestimated for dust (OPC-N2, slope = 1.6; PAS-1.108, slope = 2.7) and underestimated for welding fume (OPC-N2, slope = 0.05; PAS-1.108, slope = 0.4). The coefficient of variation (CV, precision) for OPC-N2 for all experiments was between 4.2% and 16%. These findings suggest that, given site-specific calibrations, the OPC-N2 can provide number and mass concentrations similar to the PAS-1.108 for particles larger than 1 µm.

Copyright © 2016 American Association for Aerosol Research     

Design and evaluation of a low flow personal cascade impactor

A very compact cascade impactor with 2 L/min sampling flow rate has been developed. Its dimensions are 8.5 cm L x 5.0 cm W x 11.4 cm H, and it weighs 0.27 kg, with ten impaction stages with aerodynamic cutpoints in the range of 60 nm to 9.6 μm. The top eight stages, collecting particles down to 170 nm in aerodynamic diameter, can be used as a stand-alone impactor with a portable, battery-powered pump. Particle collection efficiencies were obtained with two types of commonly used substrates, aluminum foil and glass fiber filters. Impactor cutpoints with aluminum foil substrates agree well with conventional impactor theory. The efficiency curves are sharp with minimum overlap between them. Thus, the compact impactor design does not compromise its performance, making it suitable for general purpose applications where a lower sampling flow rate provides adequate mass collection. With glass fiber filter substrates, impactor cutpoints are smaller and the efficiency curves are less steep, in particular for the last stages. Also, the collection efficiency curves do not drop to near zero at small Stokes numbers. Instead, excess particle collection efficiency of around 10% is observed for the top six stages, and becomes higher for the last four stages. This is due to the collection of particles by filtration as the impinging jets penetrate the filter substrate. Thus, using glass fiber filter substrates should generally be avoided due to the non-ideal effect on the impactor collection efficiency curves, especially for the last two stages.

Copyright © 2018 American Association for Aerosol Research     

Numerical analysis of the dynamics of aerosol inertial collection and aggregation on raindrops

A three-dimensional stochastic model is developed for predicting atmospheric aerosol collection and aggregation on the surface of a falling raindrop at its terminal velocity. Potential flow and viscous flow are assumed as the flow fields in the vicinity of the large and the small raindrops, respectively. The results show that hydrophobic coarse mode aerosols collected by either small raindrops (d_c < 100 μm) or large drops (d_c > 100 μm) form aggregations on the surfaces of drops, and accumulation mode aerosols tend to be captured by the aggregations or hydrophobic coarse particles which have been collected by the drops, and this may significantly enhance the capability of the raindrop for fine aerosol collection. When the aggregation effect is considered in the calculation, fine aerosol efficiency can be promoted by one to two orders of magnitude. Therefore, fine particle collision efficiency by raindrops is underestimated by employing the classical dynamic theory which neglects the particle aggregation effect. However, the collection efficiency of coarse particles remains almost constant with the increase in the amount of particles collected by large drops, while there is only a slight increase in efficiency by small raindrops upon increasing in particle concentration. This implies that the traditional limiting trajectory method can still be used for the calculation of coarse particle collection efficiencies by either small or large raindrops.

Copyright © 2018 American Association for Aerosol Research     

Comparison of two size-differentiating air samplers for detecting airborne swine viruses under experimental conditions

Detection and quantification of dilute viral aerosols, as encountered outside animal housing facilities, requires methods that are able to detect small numbers of viruses in large volumes of air. This study compared the performance of two size-differentiating cascade impactors; an Andersen 8-stage (ACI; 28.3 L/min) and a high volume Tisch (TCI; 1,133 L/min) to assess sampling efficiency for detecting porcine reproductive and respiratory syndrome virus (PRRSV) and influenza A virus (IAV). Samples of particles sorted by aerodynamic diameter were analyzed by quantitative polymerase chain reaction (qPCR) and collection efficiency was assessed by particle size. Collection media (minimum essential medium [MEM] and beef extract [BE]), elution technique (active versus passive), and sampling times (10, 20, and 30 min) were variables assessed for the TCI sampler. Extraction efficiency was 35% higher with BE as compared to that of MEM (p = 0.0007); active extraction technique was 19% more efficient than the passive technique (p = 0.03); time of sampling did not significantly affect the amount of virus recovered. The ACI sampler was more efficient in detecting both viruses from small and medium sized airborne particles (≤3 μm) as compared to the TCI sampler (p < 0.001). The latter sampler, however, was more efficient at IAV detection from large airborne particles (>3 μm) (p = 0.0025) indicating the potential of this sampler in detecting the presence of small amounts of viruses in aerosols under field conditions.

© 2017 American Association for Aerosol Research     

The Use of Nonlinear Acoustics as an Energy-Efficient Technique for Aerosol Removal

This article is a feasibility study on using nonlinear acoustic effects, acoustic streaming and acoustic radiation pressure, for aerosol removal in an air duct. Unlike previous research, which used acoustics solely to cause aerosol agglomeration prior to aerosol removal in traditional duct collection systems, this article considers the acoustic streaming effect, which is significant but was previously neglected. Monodispersed polystyrene spheres with diameters ranging from 0.3 to 6 μm were tested. The proposed system removed 12–20% of the submicron aerosols and 25–32% of the micron aerosols when the airflow rate was approximately 90 L/min. Acoustic streaming introduces stagnation points on the surface of the air duct and removes the aerosols by deposition. Acoustic radiation pressure causes aerosols to form agglomerates. This enhances inertial impaction and/or gravitational sedimentation, which further enhances the removal efficiency of micron aerosols. The particle-removal efficiency is proportional to the duration that the aerosols are exposed to the acoustic field. The pressure drop due to the nonlinear acoustic effects is negligible; thus, power consumption is minimal. This system has the potential to be developed into an energy-efficient technique for aerosol removal.

Copyright 2014 American Association for Aerosol Research     

Characteristics of traffic-induced fugitive dust from unpaved roads

The characteristics of fugitive dust emitted from vehicles traveling on unpaved dirt roads were measured using a suite of instruments including a real-time fugitive dust sampler. The fugitive dust sampler is formed from a combination of a large particle inlet and an optical particle spectrometer that reports particle sizes from 6 to 75 µm. The large particle inlet permits the sampling of particles up to 75 µm with only a moderate dependence of sampling efficiency on wind-speed. Measurements made with the sampler showed that particles as large as ～50 µm were suspended from vehicular movement on the dirt roads, with the mode of the fugitive dust particle number size distribution ～2 µm, while the mass distribution mode was ～7 µm. A comparison of the fugitive dust sampler measurements with those made using standard PM instruments showed that the conventional instruments have a wind-direction bias that can result in under-sampling of large particles. The current measurements suggest that particles suspended from dirt roadways are of importance for local air quality within the near-road environment.

Copyright © 2017 American Association for Aerosol Research     

Using silver as a surrogate for plutonium: An experimental study of the explosive silver aerosol source term

An experimental method is developed for the purpose of simulating plutonium aerosol source terms with conventional metals in laboratory. In this method, metal samples are aerosolized by high explosive detonation in a containment vessel. Aerosols having aerodynamic diameter (AD) less than 10 µm are then collected by a cascade impactor and analyzed by atomic absorption spectroscopy. Two sets of experiments were conducted. In the first set, five candidate metal samples (Ag, W, Sn, Ce, and V) were tested. It is found that the cumulative mass distribution of silver under certain conditions was in good agreement with that of plutonium from the Operation Roller Coaster-Double Track experiment. Thus, silver is chosen as a surrogate to simulate the plutonium aerosol source term. In the second set, silver aerosol source term was studied in detail with different test configurations. The results demonstrate that the peak of the mass-size distribution of silver is in the AD range 1.1–3.3 µm. The amount and fraction of relatively small silver aerosols decrease significantly with time due to coagulation and deposition. Interestingly, the amount of silver in aerosols could be expressed as a quadratic function of the peak detonation pressure.

© 2016 American Association for Aerosol Research     

Comparison of Particle Number Counts Measured with an Ink Jet Aerosol Generator and an Aerodynamic Particle Sizer

Aerodynamic particle sizer (APS) users typically calibrate the particle sizing capabilities, but not the counting efficiency upon which aerosol concentration results are based. Herein, comparisons were made between the counts provided by an ink jet aerosol generator (IJAG) with those measured by an APS. Near-monodisperse (geometric standard deviation of about 1.06) liquid or solid aerosols in the size range of 0.95 to 13.3 μm aerodynamic diameter (AD) generated with an IJAG were released into the inner inlet-tube of the APS in a manner that rendered APS wall and aspiration losses negligible. For most experiments, the IJAG generated 75 particles/s, which rate was maintained by the IJAG system through control of electrical pulses applied to its ink jet cartridge. For particles in the size range of 2–13.3 μm AD, the ratio of relative detection efficiency (ratio of the number of particles counted by the APS to the number reported as generated by the IJAG) was 99.3 ± 1.4%; however, for test particles between 0.95 and 2 μm AD, the relative detection efficiency was somewhat lower, but the drop off was less than about 2%. This slight drop off is likely associated with the light scattering detection approach and corresponding counting algorithm of the APS. Tests were conducted where the IJAG produced 7.0 μm AD particles at rates of 1 to 500 s^-1 and the results showed essentially a 1:1 correspondence between IJAG and APS counts. The presence of smaller-sized background particles did not affect the measured APS counts of larger-sized challenge particles.

Copyright 2014 American Association for Aerosol Research     

Rectangular Slit Atmospheric Pressure Aerodynamic Lens Aerosol Concentrator

A rectangular slit micro-aerodynamic-lens (μADL) aerosol concentrator operating at atmospheric pressure has been developed. A single stage version has shown concentration ratios of up to 40:1 for 1 μm aerosol particles while particles larger than 2 μm can be concentrated by more than 100:1 in a single stage. The design of this device has been guided by unsteady 3D CFD modeling using detached eddy simulations (DES), and has been validated experimentally using polystyrene spheres and salt crystals of known aerodynamic diameters. The pressure drop in the device does not exceed 1.5 kPa in the major flow and 0.3 kPa in the minor flow at a total flow of 10 slpm.

Copyright 2014 American Association for Aerosol Research     

Inter-comparison of low-cost sensors for measuring the mass concentration of occupational aerosols

Low-cost sensors are effective for measuring the mass concentration of ambient aerosols and second-hand smoke in homes, but their use at concentrations relevant to occupational settings has not been demonstrated. We measured the concentrations of four aerosols (salt, Arizona road dust, welding fume, and diesel exhaust) with three types of low-cost sensors (a DC1700 from Dylos and two commodity sensors from Sharp), an aerosol photometer, and reference instruments at concentrations up to 6500 µg/m³. Raw output was used to assess sensor precision and develop equations to compute mass concentrations. EPA and NIOSH protocols were used to assess the mass concentrations estimated with low-cost sensors compared to reference instruments. The detection efficiency of the DC1700 ranged from 0.04% at 0.1 µm to 108% at 5 µm, as expected, although misclassification of fine and coarse particles was observed. The raw output of the DC1700 had higher precision (lower coefficient of variation, CV = 7.4%) than that of the two sharp devices (CV = 25% and 17%), a finding attributed to differences in manufacturer calibration. Aerosol type strongly influenced sensor response, indicating the need for on-site calibration to convert sensor output to mass concentration. Once calibrated, however, the mass concentration estimated with low-cost sensors was highly correlated with that of reference instruments (R²= 0.99). These results suggest that the DC1700 and Sharp sensors are useful in estimating aerosol mass concentration for aerosols at concentrations relevant to the workplace.

© 2016 American Association for Aerosol Research     

Effect of Induced Charge on Deposition of Uniformly Charged Particles in a Pediatric Oral-Extrathoracic Airway

An in vitro study was conducted in the Alberta idealized child mouth-throat, which mimics average deposition in a set of nine 6–14-year-old subjects, to examine the enhancement of deposition of monodisperse uniformly charged particles as a result of induced electrostatic forces. A purpose-based atomizer was designed and built for generating monodisperse, uniformly charged particles. The atomizer generates droplets by jet break up under the action of capillary waves and charges them via electrostatic induction. The experiments cover different particle aerodynamic diameters (d _a = 3.6, 4.4, and 5.9 μm), at two flow rates (Q = 10 and 20 L/min), over a wide range of elementary charges per particle (0–10,000 e). The results show substantial increases in particle deposition in the present idealized pediatric mouth throat compared to neutral aerosols. Two empirical equations, as a function of Reynolds number, Stokes number, and induced charge number are introduced for the prediction of mouth-throat deposition in children, based on two different characteristic diameters of the airway.

Copyright 2014 American Association for Aerosol Research     

Clogging of Industrial High Efficiency Particulate Air (HEPA) Filters in Case of Fire: From Analytical to Large-Scale Experiments

The Institute of Radioprotection and Nuclear Safety (IRSN in French) is conducting research on the impact of a fire on the behaviour of containment devices such as high efficiency particulate air (HEPA) pleated filters for radioactive materials. This work aims to study the clogging of HEPA filters in case of fire involving realistic materials (polymers making up gloves boxes, waste treatment solvent, hydraulic oil, solid material mixtures making up a trash bin, electrical cables, and cabinets) used in nuclear facilities, from the medium to large scale. The clogging kinetics of industrial pleated HEPA filters is monitored by measuring the pressure drop of the filters and the filtered air temperature at a given filtration velocity (from 0.23 to 2.1 cm/s). Upstream HEPA filters, combustion aerosols are characterized in terms of size distribution, mass concentration, composition, and particle morphology using, respectively, a DMS500 (Cambustion^LTD), glass fiber filter sampling, and transmission electron microscope analysis of particles deposited on TEM grids. Particles emitted denote well-known fractal morphology, are composed of carbonaceous primary particles with diameters ranging from 31 nm to 48 nm and showing an high clogging efficiency. An empirical relationship has been successfully applied to the obtained results for a larger range of fuels, filtration velocities and fire conditions.

Finally, experiments have been performed on a large-scale facility, using full-scale fire scenarios (electrical cabinet, constant, and variable filtration velocity) and a reasonable agreement was observed with our empirical relationship. At this scale, particles appear to be compact, with a complex composition and diameters close to 220 nm with a lower clogging efficiency.

Copyright 2014 American Association for Aerosol Research     

Large Particle Penetration During PM10 Sampling

The objective of the present study was to characterize the performance of a federal reference method (FRM) PM₁₀ size-selective inlet using analysis methods designed to minimize uncertainty in measured sampling efficiencies for large particles such as those most often emitted from agricultural operations. The performance of an FRM PM₁₀ inlet was characterized in a wind tunnel at a wind speed of 8 km/h. Data were also collected for 20 and 25 μm particles at wind speeds of 2 and 24 km/h. Results of the present sampler evaluation compared well with those of previous studies for a similar inlet near the cutpoint, and the sampler passed the criteria required for certification as a FRM sampler when tested at 8 km/h. Sampling effectiveness values for particles with nominal diameters of 20 and 25 μm exceeded 3% for 8 and 24 km/h wind speeds in the present study and were statistically higher than both the “ideal” PM₁₀ sampler (as defined in 40 CFR 53) and the ISO (1995) standard definition of thoracic particles (p < 0.05) for 25 μm particles leading to the potential for significant sampling bias relative to the “ideal” PM₁₀ sampler when measuring large aerosols.

Copyright 2014 American Association for Aerosol Research     

Spatial aerosol flow maldistribution: A design flaw confounding the proper calibration and data interpretation of stages “0” and “1” of the Andersen eight-stage nonviable cascade impactor

We introduced monodisperse calibrant particles into an eight-stage non-viable Andersen cascade impactor (ACI) operated at 28.3 L/min and separately quantified the particle mass captured under each of the four concentric rings of nozzles on stages 0 and 1, the entry and succeeding stages of this impactor. On both stages, we found that each ring of nozzles has a particle capture efficiency behavior that differs from the others, and the fraction of calibrant particles deposited under each of the individual rings of nozzles depended on the particle size. We believe this behavior derives primarily from a radial flow velocity non-uniformity associated with recirculation zones introduced by the 110° expansion angle of the inlet cone. Because of these recirculation zones, the inertia of particles larger than about 5 µm aerodynamic diameter will cause their point-wise local concentration to differ from the concentration at the inlet entry. This concentration maldistribution continues to stage 1 primarily because of the annular collection plate at stage 0. The influence of the inlet cone aerodynamics on the performance of both stages means that the size of particles deposited on these plates will be uncertain unless the aerosol transport entering the impactor associated with calibration using monodisperse particles exactly simulates the in-use aerosol flow conditions. The degree of realism necessary in the calibration method has heretofore not been discussed in published calibrations of the ACI, introducing uncertainty in the size interpretation of the particle mass collected on stages 0 and 1 in practical applications of this impactor.

Copyright © 2017 American Association for Aerosol Research     

Development of a High-Pressure Aerodynamic Lens for Focusing Large Particles (4–10 μm) into the Aerosol Time-of-Flight Mass Spectrometer

A new aerodynamic lens system for an online aerosol time-of-flight mass spectrometer (ATOFMS) has been designed and constructed to transmit and allow the analysis of individual particles in the 4–10-μm-size range. Modeling was used to help design the lens within the bounds of ATOFMS instrumental constraints. The aerodynamic lens operates at a high inlet pressure, 3066 Pa (23 Torr), with a unique tapered relaxation region to improve large particle transmission. Every stage of the lens was tested empirically using a combination of particle deposition and light scattering experiments. The critical orifice was found to significantly impact large particle transmission, with orifices <200 μm in diameter completely suppressing large particle transmission. The addition of a virtual impactor allowed for the use of large orifices without any loss of functionality in the ATOFMS. The detection efficiency of the ATOFMS was >10% for particles from 4–10 μm with a peak efficiency of 74 ± 9% for 6-μm particles. With the extended size range provided by this inlet, the ATOFMS can now be extended to investigate single cell metabolomics.

Copyright 2014 American Association for Aerosol Research