- lauraclark849
- Oct 8
- 17 min read
Peer Review Article | Open Access | Published 8th October 2025
Qualification of the Disinfection Efficacy of an Ultraviolet Autonomous Robot for Use in Pharmaceutical Clean Rooms.
S Capper², C Cooke¹, K Capper¹*, V Hamers³, A Gravett¹, J Bright¹ | EJPPS | 303 (2025) | https://doi.org/10.37521/ejpps30307
SUMMARY
The aim of this study was to evaluate the efficacy of a UV-C disinfection robot intended for use on various cleanroom surfaces in the manufacturing facilities at AstraZeneca against a panel of different microorganisms. The studies performed used a method based on the Surface Challenge Test Method EN 13697² to evaluate the reduction in microbial viability after an inoculated surface was exposed to a with a light source emitting 254nm UV-C energy. In this surface challenge test, representative microorganisms were seeded onto coupons representing typical surfaces found in the cleanroom manufacturing areas. The inoculated coupons were treated with the UV-C light source at multiple dose-levels (mJ/cm²) to generate a death curve per organism, representing the required dose to achieve the desired log reduction as per USP<1072>¹¹ for disinfectant efficacy. The UV-C light source was tested against a panel of five American Type Culture Collection (ATCC) organisms (Pseudomonas paraeruginosa, Staphylococcus aureus, Bacillus spizizenii, Candida albicans and Aspergillus brasiliensis) and environmental isolates (Micrococcus luteus, Moraxella osloensis, Bacillus Cereus & Penicillium chrysogenum) selected from major AstraZeneca sites, which included those likely to present the greatest challenge to the effectiveness of the UV-C. The robot was qualified and shown to be suitable for use and can achieve a 3-log reduction for bacteria and 2-log reduction for spores depending on the dose applied. The dose selected by AstraZeneca was 100 mJ/cm² which represented the most practicable dose for the majority of clean room isolates with a higher dose reserved for use if adverse events for moulds or bacterial spores were recovered.
INTRODUCTION
Ultraviolet-C (UV-C) irradiation has a long-standing history as a microbicidal strategy, acting by inducing irreparable DNA and protein damage in microorganisms. In pharmaceutical clean rooms, microbial bioburden control is critical for GMP compliance and patient safety. The challenge is heightened by the prevalence of resistant organisms and the regulatory demand for evidence-based, reproducible surface disinfection. The application of autonomous UV-C robots offers promise for digitalised, chemical-free surface decontamination. However, such implementation in regulated clean rooms requires thorough qualification to demonstrate equivalence or superiority to traditional chemical disinfectants, adhering to standards such as PDA Technical Report 70⁸, EN2614-08¹, EN 13697, ² and USP <1072> ¹¹
This study was designed to:
Evaluate the disinfection efficacy of a UV-C robot on typical clean room surfaces contaminated with representative challenge organisms.
Assess if a standardised dose (100 mJ/cm²) achieves required log reductions as per regulatory standards (USP <1072>¹¹).
Determine the suitability of the robot for global AstraZeneca sites, thus avoiding the need for location-specific requalification.
Specific studies included validation of inoculation/recovery methods, assessing performance across various surface types, and providing practical recommendations for dose selection and operational integration.
The main study was a bench study to establish dose and efficacy. The scope of organisms selected covered most major types of environmental contaminants and those most seen in AstraZeneca cleanrooms using a global database of identification results from all 27 AstraZeneca sites.
The vendor provided a papers ⁴,⁹,¹⁰ which included a study commissioned at the German Hygiene Institute that showed a dose of 100 mJ/cm² was effective at gaining a good overall log reduction for most isolates and this therefore was chosen as a target dose for the study. The study also showed the vegetative organisms were much more sensitive, so lower doses would be needed for those to get a good range of results for a death curve. Microorganisms recommended in USP 35 <1072>¹¹ were also included. A paper by Freundt, 2016⁵ details the resistance of Micrococcus sp to UV light, it states that Micrococcus luteus and Micrococcus radiophilus are bacteria known for their resistance to high UV levels. Micrococcus luteus is one of the top occurring isolates in the AstraZeneca cleanrooms and so for this reason was selected as one of the environmental isolates. It is also known that organisms that produce bacterial and fungal spores are harder to kill than vegetative cells. For this reason spore suspensions were chosen for the UV study for the ones recommended in USP <1072>¹¹ and environmental Bacillus sp and mould isolates. For robustness more than one microbiologist performed studies on different days to remove the factor of operator difference in the test methods. A further paper by Fuchs, 2022⁶ provided some examples of how to set up the experiment. They found that the distance between the coupon and the light was critical to standardise. A standard set up was established for a standing robot tower that allowed the light to shine directly on to the coupon and gave remote access to switch on the light and measure the dose and time. The distance from the coupon was standardised in the same way as this study. A remote light switch was used to switch on before exposing the coupons.
METHODS
The qualification study followed a two-phase approach: (1) pre-work laboratory validation, and (2) main qualification testing, both guided by EN 13697²and USP <1072>¹¹ frameworks.
Pre-Work Studies
The initial phase optimised coupon inoculation, recovery, and enumeration methods for four compendial organisms, Staphylococcus aureus, Candida albicans, Pseudomonas paraeruginosa, and Escherichia coli. Both spread plate and alternative enumeration methods were compared, with Bovine Serum Albumin (BSA) evaluated as a desiccation protectant. Coupons were prepared, inoculated, dried under controlled conditions, and organism recovery was quantified. Representative microorganisms were seeded onto vinyl coupons. The coupons were 3cm² swatches of material. Microbial recovery was evaluated by rinsing the coupons followed by an enumeration of the rinse. The purpose of the study was to generate ruggedness and robust data as well as best practices around the use of coupons in the main study. The study was executed over multiple days to allow for testing of the range of techniques with the 4 different organisms. For this experimental programme, samples were prepared via the pour plate, spread plate and filtration methods. The technique which demonstrated the best recovery was selected as the spread plate and taken forward into the main study. Literature for disinfection studies² suggests that Gram negative organisms require the albumin to avoid desiccation during the pre-study. However, it was found that the albumin caused lower recovery post UV treatment so was not used in the main study. It was considered vinyl was the worst case for the prework as the surface was less smooth than steel and glass. No in-house isolates were used during this pre work.
The UV robot set up and dose measurement was also established. A UVC tower replica, which included eight low vapor pressure mercury lamps at 180 Watts each with reflector panels behind each lamp, was set up and a webcam was used in the room with a remote Wi-Fi switch used to turn the light on and off from outside the controlled access room, Figure 1a and 1b.
Measures the dose of UV received by the coupon and this is tracked by the webcam. Battery powered; microprocessor based. Provides an accurate numerical display of the dose, mJ/cm².
Main Study
Using the optimised protocol, test coupons seeded with challenge organisms, comprising compendial strains and frequently encountered AstraZeneca environmental isolates, were exposed to the robot’s UV-C source at specified doses (ranging from 0 to 500 mJ/cm²). The panel included vegetative bacteria, bacterial spores, and fungal spores, as detailed in Table 1.
Table 1. Microorganisms included in the main study
Microorganism and strain | Characteristics & Reasoning for Study Inclusion |
Staphylococcus aureus (ATCC 6538) | Gram-positive Cocci, human skin. USP<1072> |
Pseudomonas paraeruginosa (ATCC 9027) | Gram-negative rod, water sources USP<1072> |
Bacillus spizizenii spores (ATCC 6633) | Gram-positive bacillus, environmental, resistant to disinfection USP<1072> |
Candida albicans (ATCC 10231) | Yeast, human skin USP<1072> |
Aspergillus brasiliensis spores (ATCC 16404) | Mould, environmental, resistant to disinfection USP<1072> |
Moraxella osloensis environmental isolate | Gram-negative rod, human skin, high occurring Gram-negative isolate from clean rooms. |
Micrococcus luteus environmental isolate | Gram-positive Cocci, human skin, high cleanroom occurring isolate and UV resistance. |
Penicillium chrysogenum environmental isolate | Mould, environmental, resistant to disinfection, high occurring fungal isolate in transfer areas. |
Bacillus cereus environmental isolate | Gram-positive bacillus, environmental, resistant to disinfection and objectionable. |
Preparation
Prior to experimental use, test coupons were immersed in Decon 90® cleaning solution for 24 hours to ensure thorough decontamination. Coupons were subsequently rinsed with sterile water to eliminate residual cleaning agent and dried with a sterile wipe. Enumeration of in-house microbial isolates was carried out according to established laboratory protocols to determine appropriate dilution factors for inoculation. Aseptic technique was adopted throughout.
Test Method
The main disinfection efficacy studies were conducted utilising the optimised protocol established during the pre-work phase, with procedures aligned to the Surface Challenge Test Method EN 13697². Representative microorganisms were inoculated onto test coupons composed of glass, vinyl, and stainless steel. After inoculation, coupons were subjected to UV-C exposure at a range of precisely measured dose levels (in mJ/cm²) in order to construct microorganism-specific death curves and to determine the minimum effective dose necessary to achieve the targeted log reduction in viable counts.
For each test, coupons were individually inoculated with suspensions of either compendial ATCC strains (Pseudomonas paraeruginosa, Staphylococcus aureus, Bacillus spizizenii, Candida albicans, Aspergillus brasiliensis) or environmental isolates (Moraxella osloensis, Micrococcus luteus, Penicillium chrysogenum, Bacillus cereus spores) selected based on their prevalence and relevance to pharmaceutical clean room environments. Environmental isolates were prepared from stock cultures, using serial dilution and plate count verification to adjust inoculum concentration to approximately 10⁶ CFU per 0.01 mL. The diluted suspensions were prepared less than two hours prior to use to preserve viability.
Each coupon batch included both positive and negative controls. Positive control coupons received an inoculum but were not exposed to UV-C, whereas negative controls were un-inoculated. All test, control, and negative coupons were cleaned with Decon 90® and dried with a sterile wipe prior to inoculation.
Following inoculation, coupons were dried as shown in Figure 3, transported in sterile, closed containers to the UV-C exposure facility. Each organism and material combination were tested in duplicate across all dose levels; for each set, two test coupons, two positive control coupons, and one negative control coupon were used.
Post-exposure, residual microorganisms were recovered from each coupon by aseptically transferring into 5 mL of sterile buffered sodium chloride peptone (BSCP) containing sterile glass beads, followed by vortex mixing for 2 minutes. The fluid was serially diluted and plated using the spread plate method onto tryptic soy agar. Bacterial plates were incubated at 30–35°C for three days, while moulds were incubated at 20–25°C for five days. Colony counts from each plate were recorded and corrected for dilution, and the log reduction in microbial viability was calculated by comparison to corresponding positive control coupons. Colony counts were recorded to determine final CFU values. This was to enable a target 3-log reduction for vegetative organisms and a target 2-log reduction for spores shown in Table 2.
Table 2. UV doses to be administered per organism type in mJ/cm²
Vegetative orgs | Bacterial spores | Mould spores |
0 | 0 | 0 |
5 | 20 | 20 |
10 | 40 | 40 |
15 | 60 | 80 |
20 | 80 | 100 |
25 | 100 | 200 |
50 | 200 | 500 |
100 | 700 |
For each experiment, two inoculated test coupons per material (glass, vinyl, and steel; six coupons in total) were aseptically removed from containment and placed upright on a shelf at a standardised distance of 2 metres from the UV-C light source to ensure uniform and accurate dosing. Coupons were oriented vertically to maximise exposure. Positive controls (two per material) and negative control coupons (one per material; nine coupons in total) were similarly removed and placed in BSCP in the same environment protected from UV exposure. Figure 4.
UV dose was actively monitored during exposure using the UVKEY® dosimeter (Figure 2), positioned alongside the coupons, providing real-time dose feedback. UV intensity was additionally quantified using an ILT2400® light meter. The progress of dose accumulation was observed remotely via a webcam, allowing the experimenter to terminate exposure precisely upon reaching the target dose.
Before irradiation, both measurement devices were reset to zero, and all laboratory personnel exited the room. The UV-C source was activated remotely, with a timer initiated to track exposure duration. Upon delivery of the defined UV dose, the light source and timer were switched off.
Immediately post-exposure, all test, positive control, and negative control coupons were transferred into 5 mL sterile buffered sodium chloride peptone (BSCP) with sterile glass beads, inoculated side down, to facilitate the recovery of microorganisms. This recovery procedure was conducted identically for both exposed and unexposed control coupons.
To recover organisms from the surface of each coupon, 5 mL of buffered sodium chloride peptone (BSCP) extract containing the coupon and glass beads was vortexed for 2 minutes. Serial tenfold dilutions were prepared by transferring 1 mL of the extract into 9 mL sterile BSCP, extending through to a final dilution of 10⁻⁶. For enumeration, 0.1 mL aliquots from the 10⁻¹ to 10⁻⁶ dilutions were plated in duplicate onto tryptic soy agar (TSA) plates for both positive control and inoculated test coupons. Colony counts were adjusted for the plated volume during calculation of colony-forming units (CFU). Negative control coupons underwent the same procedure: each was rinsed with 5 mL of sterile BSCP, subjected to serial dilution, and plated as described for test coupons, to confirm sterility of the process.
The effectiveness of UV-C disinfection was assessed by calculating the log reduction in viable microorganism counts, using the following equation:
The log reduction was calculated as follows:
R = P1 – P2
Where,
R = Log₁₀ Reduction of a challenge organism by the UV
P1 = Log₁₀ Average cell density recovered from the unexposed coupon (Positive Control Count).
P2 = Log₁₀ Average cell density recovered from the exposed Test Coupon
For each test organism, survival data across UV-C dose levels were plotted to generate organism-specific death curves. Figure 7.
Acceptance criteria applied to the study
Following the principles of USP <1072>¹¹, the UV surface test will be considered effective against vegetative test organisms if the cell density of viable microorganisms is reduced by at least 3 logs from the initial concentration (Positive Control Count) USP <1072>¹¹.
The UV will be considered sporicidal and effective against fungi if the concentration of viable bacteria spores/fungi is reduced by at least 2 logs from the initial cell density (Positive Control Count). USP <1072>¹¹.
Positive Control Count: The viable microbial count must yield a target of approximately 10⁶ CFU for each compendial challenge organism and 10⁶ for environmental isolates.
Negative Control Coupons: The viable microbial count must show 0 CFU.
RESULTS
The results shown here are for the target 100 mJ/cm² dose.
Table 3. 100 mJ/cm² dose results
Organism Name | Organism Type | Surface Type (worst case) | Target Log Reduction | Log Reduction achieved. |
Bacillus spizizenii Spores | Bacterial Spores | Vinyl | 2-log | 1.2-2.9 |
Bacillus cereus Spores (EI) | Bacterial Spores | Vinyl | 2-log | 1.6->2.0 |
Aspergillus brasiliensis spores | Fungal Spores | Glass | 2-log | 0.8->1.8 |
Penicillium chrysogenum Spores (EI) | Fungal Spores | Glass | 2-log | 0.5-2.4 |
Candida albicans | Yeast | All | 3-log | >3.0 |
Staphylococcus aureus | Gram Positive Cocci | All | 3-log | >3.0 |
Micrococcus luteus (EI) | Gram Positive Cocci | Vinyl | 3-log | 2.8->3.0 |
Pseudomonas paraeruginosa | Gram negative rods | All | 3-log | >3.0 |
Moraxella osloensis (EI) | Gram negative coccobacillus | All | 3-log | >3.0 |
Table 4 below shows the dose required to gain the target log reduction on all surfaces for each organism. Organisms are grouped in type.
Table 4. Dose required to hit the acceptance criteria on all surfaces.
Organism Name | Organism Type | Dose Required for Target 2 or 3-Log Reduction on all Surfaces mJ/cm2 |
Bacillus spizizenii Spores | Bacterial Spores | >200* |
Bacillus cereus Spores (EI) | Bacterial Spores | 200 |
Aspergillus brasiliensis spores | Fungal Spores | 500 |
Penicillium chrysogenum Spores (EI) | Fungal Spores | 200 |
Candida albicans | Yeast | 50 |
Staphylococcus aureus | Gram Positive Cocci | 100 |
Micrococcus luteus (EI) | Gram Positive Cocci | 150 |
Pseudomonas paraeruginosa | Gram negative rods | 100 |
Moraxella osloensis (EI) | Gram negative rods | 40 |
* 200mJ/cm² on glass gave a log reduction of 2.8 and steel 3. On vinyl it gave a log reduction 1.54 Higher doses were not performed but on balance across all 3 surfaces 200mJ was sufficient.
For each tested microorganism, the log reduction (log₁₀ CFU) was plotted in Figures 5 and 6 against the applied UV-C dose 100 mJ/cm², enabling the visualisation of death curves across all organisms and surfaces. When the UV gives a total kill the log reduction is infinite in truth, but a value of 3 is shown with a solid upward arrow to represent >3. Otherwise, the plotting symbol is an open circle appearing at the calculated value of log reduction according to the rules described in "Construction of Death Curves" below.
The efficacy of the UV-C surface disinfection was assessed according to USP <1072>¹¹ criteria: an effective result against vegetative organisms was defined as a minimum 3-log₁₀ reduction in viable count relative to the initial (positive control) concentration, while a minimum 2-log₁₀ reduction was required for spores and fungi. At a dose of 100 mJ/cm², all vegetative bacterial and yeast challenge organisms achieved at least a 3-log₁₀ reduction, with the exception of Micrococcus luteus, which attained between 2.8 and 3 -log₁₀ reduction. The 100mJ/cm² on glass gave a total kill, so given a log reduction of 3. On steel it gave a log reduction of 3.1, however on vinyl it gave a log reduction of 2.8. For sporicidal and antifungal activity, for the 12 combinations of organisms and surface only 4 achieved the threshold log reduction of 2. Figure 6. Indicating a higher dose is needed to meet that but there was still an effective reduction.
The intended positive control counts were approximately 10⁸ CFU for compendial organisms and 10⁶ CFU for environmental isolates; however, these high inocula were not always attainable with small coupon surfaces. Experimental optimisation demonstrated that recoveries of 10⁶ CFU per coupon were sufficient for robust log reduction assessments. All negative control coupons showed no detectable microbial growth, confirming procedural disinfection
Data Handling, Positive Control Calculation, and Statistical Methods
For experiments involving repeated runs or multi-day testing, only data from contemporaneous sets, defined as those tested head-to-head using the same batch of positive controls, were used for comparative analysis and graphical representation. For instance, if UV doses of 20, 50, and 200 mJ/cm² were tested on a given day alongside one set of positive controls, those controls served as the reference for these doses only; any later experiments with different controls (e.g., a 100 mJ/cm² run on a different occasion) were compared to their own contemporaneous positive control set.
All results were plotted on a logarithmic vertical axis, ensuring comparability of log reductions across the tested range (e.g., 1-log reduction between 1,000 and 100 CFU equates graphically to 1-log reduction between 100 and 10 CFU).
Construction of Death Curves
To ensure reproducibility and consistency in the generation of microbial inactivation (death) curves, a standardised approach was adopted for data selection and processing, particularly in handling positive control values and replicate variability. The following criteria were established:
1. Dilution Selection: For both positive controls and UV-exposed coupons, only counts from the most concentrated dilution in which all non-missing values from two coupons (each with two technical replicates) were less than 300 CFU were included in calculations. The arithmetic mean of these counts was computed and multiplied by the appropriate dilution factor (e.g., 1 for neat, 10 for -1 dilution) to determine the final value.
2. Exclusion of Repeats: Repeat experiments were excluded from the death curve analysis to avoid duplication and potential bias.
3. Handling of Complete Inactivation: Instances in which UV-treated coupons yielded an average CFU count of zero (indicating complete inactivation) were expressed as a fixed log reduction value of 3, representing the maximum measurable reduction under the experimental conditions.
By adhering to these predetermined rules, inherent variability common in microbiological assays was minimised, enabling reliable comparison of log-reduction values across UV-C doses and tested surfaces.
The death curves are shown in Figure 7.
DISCUSSION
During the feasibility phase, robust recovery of all four tested organisms was observed using the coupon protocol. Dilutions at 2.2 x 10³ CFU and above were consistently reported as too numerous to count (TNTC), while countable recoveries were obtained at 2.2 x 10² CFU and below. Comparison of three enumeration methods indicated that the spread plate method yielded the highest and most reliable organism recovery, thereby being selected for further studies because of its simplicity and effectiveness. Notably, bovine serum albumin supplementation was not required in the main study, as it was found to impede UV-C penetration and thus potentially interfere with microbial inactivation.³ While previous literature³ suggested that bovine serum albumin may help preserve Gram-negative viability during drying, its exclusion accelerated drying times without having an adverse impact on recovery for the organisms tested. These findings are in line with Lourenço et al.⁷, who observed that higher BSA concentrations could reduce the effectiveness of light-activated antimicrobial surfaces.
The main efficacy study demonstrated that UV-C treatment at 100 mJ/cm² consistently achieved at least a 3-log reduction in vegetative organisms from their initial concentration, with the exception of a single Micrococcus luteus replicate, which showed a 2.6-log reduction. This is anticipated due to the inherently challenging nature of microbiological methods. Factors such as surface irregularities, higher challenge inoculum, and variable microbial adhesion can contribute to inconsistent results across replicates. In this study, the observed variability, principally isolated to Micrococcus luteus, a known UV-resistant organism, was not sufficient to compromise overall conclusions regarding UV-C robot efficacy, particularly in the context of its use as part of a holistic disinfection program.
For bacterial and fungal spores, while 2-log reductions were observed for some organism/surface combinations, others did not consistently meet this criterion at 100 mJ/cm². These findings suggest that, for remediation following detection of spore-forming organisms in environmental monitoring, a UV-C dose higher than 100 mJ/cm² may be required. Data from dose-response experiments and death curves indicate that doses above 200 mJ/cm² for Bacillus spores and 500 mJ/cm² for fungal spores may be warranted, ideally in conjunction with manual sporicidal disinfection.
Glass, steel, and vinyl were selected as representative cleanroom surfaces for efficacy testing. Results across these materials demonstrated broadly comparable log reductions, with the UV-C robot performing effectively on each type. However, a slight advantage in log reduction was observed on stainless steel surfaces, particularly for spore-forming organisms. Statistical analysis indicated that differences were not generally significant unless specific outlier data (e.g., from Candida albicans) were excluded, suggesting steel may be modestly more amenable to UV inactivation. Further research using additional surface types may be warranted to substantiate and generalise these observations.
CONCLUSIONS AND RECOMMENDATIONS
The autonomous UV-C robot has a suitable and effective efficacy for use in pharmaceutical clean rooms, especially for the routine reduction of vegetative bioburden at 100 mJ/cm². Its efficacy against spores is dose-dependent; thus, situational adjustments to higher doses (≥200 mJ/cm² for Bacillus spores; ≥500 mJ/cm² for resilient moulds) are recommended when environmental monitoring indicates such challenges. Key recommendations include further replication to improve statistical power and post-implementation environmental trend analysis. The UV-C robot should be used adjunctively, not as a sole disinfection method, in clean rooms, with clear protocols for addressing its known limitations (surface occlusion, variable spore reduction).
Acknowledgements
Blue Ocean Robotics, Laura Guardi, Lewis Davies, Grace Tomlinson, Etienne Wijnands, Emily Butterworth and Lisa Davis.
References
1. ASTM (2015) E2614-08 Guide for evaluation of Cleanroom Disinfectants. Available at: https://store.astm.org/e2614-08.html. Accessed on 2/2/2025
2. CEN (European Committee for Standardization), 2019. EN 13697:2015+A1:2019 Chemical disinfectants and antiseptics – Quantitative non-porous surface test for the evaluation of bactericidal and/or fungicidal activity of chemical disinfectants used in food, industrial, domestic and institutional areas – Test method and requirements (phase 2, step 2). Brussels: CEN. Accessed on 2/3/2025
3. Coughenour, C.A. (2009) ‘An evaluation of methicillin resistant Staphylococcus aureus survival on five environmental surfaces under two different humidities, with and without the addition of bovine serum albumin’. Available at: https://doi.org/10.34917/2073708. Accessed on 12/4/2025
4. Edwards-Jones, V., 2019. Assessment of UVD robot against reduction of multi-drug resistant Klebsiella pneumoniae, Acinetobacter baumanii and Clostridium difficile on surfaces. Clinical Director, Melbec Microbiology Ltd and Independent Microbiology Consultant, Essential Microbiology Ltd.
5. Freundt, E. (2016) Protective Qualities of UV-resistant Bacteria, Academia.edu. Acta Spartae. Available at: https://www.academia.edu/download/122318051/Goldberg_AS_2016_A.pdf . Accessed: 10 June 2025.
6. Fuchs, F.M. et al. (2022) ‘Characterization of a robot-assisted UV-C disinfection for the inactivation of surface-associated microorganisms and viruses’, Journal of Photochemistry and Photobiology, 11, p. 100123. Available at: https://doi.org/10.1016/j.jpap.2022.100123. Accessed on 10/6/2025
7. Lourenço, C. et al. (2018) ‘Effects of bovine serum albumin on light activated antimicrobial surfaces’, RSC Advances, 8(60), pp. 34252–34258. Available at: https://doi.org/10.1039/c8ra04361b. Accessed on 2/2/2025
8. Parenteral Drug Association. (2015) Technical Report 70. Fundamentals of Cleaning and Disinfection Programs for Aseptic Manufacturing Facilities. Available at: https://store.pda.org/TableOfContents/TR70_TOC.pdf Accessed on 02/02/2025
9. Pfleger, S.G., Haertel, M.E.M. and Plentz, P.D.M. (2025) ‘UV‐C Disinfection Robots: A Systematic Review’, Journal of Field Robotics [Preprint]. Available at: https://doi.org/10.1002/rob.22555. Accessed on 02/02/2025
10. Rager, A.-M. (2023) Development of a validation model for the safe implementation of UV disinfection by a robot, www.hyginst.de. Hygiene-Institut Schubert – Teil der DERNBACHER GRUPPE KATHARINA KASPER DGKK Dienstleistung GmbH – Rheinstraße 9 – 56428 Dernbach (Westerwald): Hygieneinstitut, pp. 1–33.
11. United States Pharmacopeia. (2024) Chapter <1072> Disinfectants and Antiseptics. Available at: https://online.uspnf.com/uspnf. Accessed on 02/02/2025
Authors
S Capper², C Cooke¹, K Capper¹*, V Hamers³, A Gravett¹, J Bright¹
¹ Global Product Development, Pharmaceutical Technology & Development, Operations, AstraZeneca, Macclesfield UK
² Quality Control, Operations, AstraZeneca, Macclesfield UK
³ Quality Assurance, Operations, AstraZeneca, Nijmegen, The Netherlands
Corresponding Author: K Capper¹*
¹*Global Product Development,
Pharmaceutical Technology & Development,
Operations,
AstraZeneca, Macclesfield UK
Email: karen.capper@astrazeneca.com
Phone: +44 7384 438531