Laboratory Research

Copper is effective against a broad spectrum of microorganisms

Laboratory research was the first to demonstrate that copper surfaces kill a broad spectrum of disease causing bacteria, viruses, and fungi.  A few studies taken from this large body of evidence are highlighted here to give an idea of the underlying science that supports the potential value of antimicrobial copper in human health.   

Why is laboratory research so important? Laboratory tests are conduced under controlled conditions with known and fully characterized strains of specific microbes usually obtained from trusted sources, such as the American Type Culture Collection (ATTC).  The ATTC collects, stores, and distributes standard reference microorganisms for research and development purposes.  Laboratory testing is done using defined protocols that can be reproduced in other laboratories. 


A typical test of the antimicrobial activity is carried out as follows.  Test samples are sanitized prior to inoculation with a measured amount of the bacterial strain.  The level of viable bacteria on the sample is then determined over time.  At each time point, the sample is washed to remove the bacteria and the wash solution cultured to determine the number of surviving bacteria.  In the laboratory tests, the inoculum concentrations were typically over 10 million colony forming units (or 10,000,000 CFUs) per square cm.  This is orders of magnitude higher than found on the typical hospital surface, which normally range from roughly 10 to 100 CFUs per square cm.

In clinical trials, samples are collected by swabbing a specific surface area of an item in a healthcare facility.  These samples contain several types of bacteria, and, in all cases, the isolates are distinct from the characterized species in the ATTC collection.  There is also no prior knowledge of how or when the surface became contaminated, what types of bacteria are on the contaminated surface, and if and when the surface was last cleaned and/or sanitized.  The samples may also contain traces of cleaning solutions, oil from hands, and contaminants from sanitizing chemicals.  Thus clinical trials conducted in hospitals are stochastic in nature.  Nevertheless, they are needed because they reflect real world conditions.  


Laboratory tests remove this randomness because of their highly controlled nature and more definitive conclusions can be drawn.  This section summarizes some laboratory test results.  It not intended to replace the large body of laboratory research that has been published about antimicrobial copper.


Time (mins)

Screen Shot 2019-10-07 at 12.12.png

First Laboratory Tests: E. coli O157:H7

In the first test related to the antimicrobial copper project, over twenty copper alloys covering a broad range of compositions, were exposed to one strain of bacteria, E. coli O157:H7 in Professor C. W. Keevil's laboratory at the University of Southampton, in the UK.  This bacteria, E. coli O157:H7, has been involved in multiple food recalls often resulting in outbreaks of severe gastrointestinal illnesses, kidney failure and even death.  All of the copper alloys were effective in killing the bacteria, over varying time intervals, which correlated with copper content, with the the highest copper content acting the fastest.  As expected, the stainless steel experimental control exhibited no meaningful efficacy.  As an example, a graph of the the the time to kill the bacteria when exposed to a series of copper nickel alloy, ranging in copper content from 97% to 67% is shown below.  The alloy with the highest copper content exhibited shortest time to kill, and this time to kill increased as copper content decreased.

1.  S. A. Wilks, H. Michels and C. W. Keevil, The Survival of  Escherichia coli O157 on a range of metal surfaces, International Journal of Food Microbiology, 105, pp. 445-454, 2005

A visual illustration of the killing power of copper, with stainless steel serving as a control, is shown below. These  images were taken with an Epifluorescence Microscope in the laboratory of Professor C. W Keevil of the University of Southampton in England.  The red dots are indicative of live E. coli, which decrease on the copper sample from 31.4 million CFUs at Time = 0, to 1.6 million CFUs at 30 minutes, 2,740 CFUs at 60 minutes and <0.1% CFUs at 120 minutes.  Stainless steel started at Time = 0 with 31.3 million CFUs and this decreased to 26.9 million CFUs at 30 minutes, 25.9 million CFUs at 60 minutes and 21.1 CFUs at 120 minutes.  The initial reduction seen on stainless steel is a result of evaporation, but the bacteria continue to survive are viable for a prolonged period.  For example, test show little reduction in the number of viable bacteria on the stainless steel sample after 28 days, when the test was terminated.

Unpublished results, C. W. Keevil

Listeria monocytogenenes

The Centers for Disease Control and Prevention indicates that those individuals infected by Listeria have a very high hospitalization rate (90%) and also a rather high mortality rate (20%). This food-borne pathogen is especially dangerous to pregnant women, the elderly, and those who are immune compromised.  The graph show below illustrates the reduction of live Listeria monocytogenenes when placed on a series of copper alloys ranging in copper content from 100% to 65%.  The time to complete kill was inversely related to the copper concentration of the alloy.  The 65% copper alloy, which was the lowest concentration tested, took 90 minutes, with the other copper alloys taking between 60 to 75 minutes. Stainless steel, which contains no copper, served as the experimental control.  It showed no killing over the course of the experiment. The result is similar to what was seen in tests of E. coli .

Screen Shot 2019-10-07 at 3.24.35 PM.png

2.  S. A. Wilks, H. T. Michels and C. W. Keevil, Survival of  Listeria monocytogenes Scott A on metal surfaces: Implications for cross-contamination, International Journal of Food Microbiology, 111, pp. 93-98, September 2006 

Methicillin-Resistant staphylococcus aureus (MRSA)

Staphylococci bacteria, or staph, are ubiquitous and are present on the skin of many people.  Staph is easily transmitted between individuals, from patients to healthcare workers and to others.  Those with a break in the skin, an open wound, or with compromised immune systems are at risk.  The antibiotic-resistant strain of Staph, MRSA,  is one of the so-called Hospital 'Super Bugs'.  Its ability to infect is particularly aggressive making it a very serious contaminant in the healthcare environment.   MRSA infections are responsible for about 126,000 hospitalizations each year in the US. MRSA has also jumped to be a communiuty acquired infection  has become a problem in professional and schools sports.  

3.  M. J. Kuehnert, H. A. Hill, B. A. Kupronis, J. I. Tokas, S.  L. Solomon and D. B. Jerigan, Methicillin-resistant Staphylococcus aureus Hospitalizations, Uniteded StatesEmerg. Infect. Dis., 11(6), 868-872, 2011

An epidemic variant strain of MRSA, called EMRSA, that caused outbreaks in several hospitals was recognized as a distinct strain in 1981 and identified as EMRSA-1.   Noyes et al. (2006) evaluated survival of three strains MRSA and two epidemic strains, EMRSA- 1 and EMRSA-16, following exposure to two copper alloys, C197 (99% Cu) and brass C240 (80% Cu - 20% Zn).  (Note:  The figure legend should read ERMSA-16 and rather than ERMSA-1'.)  Exposure to stainless steel served as the experimental control; no meaningful loss in viability was observed.  In contrast, exposure to C197 (99%Cu) caused complete killing of all three strains in 90 minutes of less: MRSA, EMRSA-1 and EMRSA-16. 


MRSA is killed on C24O (80%Cu) but more slowly.  Less than 2 logs or 100-fold reduction in viability is observed at 360 minutes. While a 2 log or 99% reduction is meaningful, it will take a longer time to achieve complete killing of MRSA.  In contrast, C240 achieved complete killing of both epidemic strains, EMRSA-1 and EMRSA-16, in 270 minutes. The reduced rate of killing by C240 (80% Cu) compared to C197 (100% Cu) is thought to be due to its lower copper content.  It should be noted that both EMRSA stains are more sensitive to copper than MRSA when challenged with the alloy of lower copper content, a characteristic that should help in eradicating them when an outbreak occurs.

Screen Shot 2019-10-08 at 8.41.21 AM.png

Noyces et al. (2006) also investigated the effect on inoculum size, or number of bacteria, on the time to a complete kill.  As can be seen below,  as the amount of inoculum is reduce from 10 million down to 100 CFUs in six steps, the time for complete kill of EMRA-16 is reduced from 90 minutes to 30 minutes.  The practical implication of this experiment is that copper alloy surfaces will achieve a complete kill in a short time in hospitals in comparison to laboratory experiments, because the amount of bacterial contamination is much lower in healthcare settings.  However, even a small amount of bacteria has the potential to cause infections.

Screen Shot 2019-10-08 at 8.43.06 AM.png

4.  J. O. Noyce, H. Michels and C. W. Keevil, Potential use of copper surfaces to reduce survival of epidemic methicillin-resistant Staphylococcuaureus in the healthcare environment, Journal of Hospital Infection, 63 (3) 289-297, 2006 

The effect of alloy copper content on the time for a complete kill was tested using MRSA.  The alloys in this experiment are C197 (99% Cu), C240 (80% Cu), C770 (55% Cu), and S304 stainless steel was used as the experimental control (0% Cu).  As shown in the figure below, C197 (99% Cu) achieved complete killing in 90 minutes and C240 (80% Cu) in 270 minutes.  C770 (55% Cu) displayed only a 3 log reduction (99.9%) after 6 hours of exposure.  The results clearly show that the higher the copper concentration of the alloy the more rapid it kills MRSA.  This is consistent with finding shown above for E. coli and Listeria.

Screen Shot 2019-10-06 at 5.27.07 PM.png

5.  H. T. Michels, J. P. Noyce, S. A. Wilks and C. W. Keevil, Copper Alloys for Human Infectious Disease Control, Copper for the 21st Century,        Materials Science & Technology 2005 (MS&T’05) Conference, Pittsburgh, PA, ASM, ACerS, AIST, AWS, TMS, September 25-28, 2005            (ISSN: 1546-2498) 

Vancomycin-Resistant Enterococci (VRE)

VRE is another serious antibiotic - resistant 'super bug' that causes difficult-to-treat infections.  Approximately one-third of the enterococcal infections acquired by ICU patients are found to be caused by VRE.  VRE is primarily transferred by touching between patients, healthcare workers and surfaces.  Thus, copper has the potential be very effective in killing this disease-causing bacteria and reducing infection rates.  To test this hypothesis, a number of copper alloys of different copper concentrations were inoculated with vancomycin-resistant Entercoccus faecalis and vancomycin-resistant Entercoccus faecium strains obtained from clinical isolates as well as from the NCTC (National Collection of Type Cultures).  Stainless steel was again used as the experimental control.  The results, shown below are only for vancomycin-resistant Entercoccus faecium.  The first graph in the upper right shows the data from the NCTC strain, and the other graphs are from various clinical isolates.  As you can see, the stainless steel (solid circles) produce little to no loss in viability.  The trend observed in the copper alloys C110 (100% Cu), C510 (95% Cu), C706 (70% Cu), C752 (65% Cu) and C280 (60% Cu) is clear.  The higher copper content of the alloy correlates with faster killing time, with minor some exceptions.  The results for vancomycin-resistant Entercoccus  faecalis, which are not shown, are quite similar.  In summary, copper is effective in killing VRE.

Screen Shot 2019-10-08 at 4.01.51 PM.png

6. S. L. Warnes, S. M. Green, H. T. Michels and C. W. Keevil, Biocidal Efficacy of Copper Alloys against Pathogenic Enterococci Involves Degradation of Genomic and Plasmid DNAs, Applied and Environmental Microbiology, pp. 5390-5401, Vol. 76, No. 16, Aug. 2010 

Clostridium difficile

Clostridium difficile is an anaerobic spore forming bacteria that is highly resilient and particularly infects patients taking broad spectrum antibiotics.  When stressed, Cdifficile responds by forming spores as a survival mechanism.  Cdifficile spores are not readily killed by hospital grade disinfectants, including quaternary ammonia.  However, it is easier to kill vegetative Cdifficile, as can be seen in in the graph shown below.  This plot is a simplified version of Figure 3 from the original paper.  This version only shows C110 (100% Cu), C260 brass (70% Cu) and the experimental control, S304 stainless steel.  The paper can be viewed by opening the PDF shown below.  In Figure 3 in the paper , you will see a more detailed plot, that includes the other three copper alloys, C510 (95% Cu), C706 (90% Cu) and C752 (83% Cu).  These three alloys are very similar to C110 (100% Cu) and achieve over a 6 log drop in viability within 6 hours when tested against Cdifficile.  Thus, copper is effective against Cdifficile vegetative cells in copper alloys containing  at least 70% Cu within 6 hours.  Complete killing can take 48 hours, which is a prolonged time to kill this organism.  Figure 4 in the same paper shows similar efficacy against A. fumigatus, spores, at less than 24 hours.  However,  ~1/2 log resistance can be seen between 24 and 48 hours in comparison compared to Figure 3.  Thus, as expected, the spores are more resistant than vegetative Cdifficile.  In summary, while  copper is somewhat effective against Cdifficile, additional studies are needed to better understand how to better understand and reduce the time it takes copper to kill Cdifficile.

7.  L. Weaver, H. T. Michels, and C. W. Keevil, Survival of Clostridium difficile on copper and steel: futuristic options for hospital hygiene, Journal of Hospital Infection, 68 (2)145-151, 200808


Fungal isolates of Candida albicans, and  several spore forming fungi, including Aspergillus niger, A. flavus, A. fumigatus, Penicillium chrysogenum, plus Fusarium culmonium, F. oxysporium and F. solani were exposed to copper and aluminum.  Exposure for 24 hours to copper resulted in a complete killing of Candida albicans, P. chrysogenum and all Fusarium species, but no change in their viability was seen on aluminum, the experimental control.  Complete die off of A. flavus was seen at 120 hours, and it took 144 hours for A. fumigatus to die on copper.  In contrast, it took longer for copper to be effective against most Aspergillus species.  Aspergillus niger is quite robust, showing no significant killing even after 576 hours. 

However, the image presented below indicates that copper does have some inhibition effect on the growth of Aspergillus niger.   Spore suspensions were spread over both the copper and aluminum samples and incubated at ~22 degrees C for 10 days.  Note that growth is visible on the aluminum sample but not on the copper.  Also observe the region around the edges of the copper in which growth is absent, referred to as a zone of inhibition.  In summary, copper is antifungal after spores germinate, but it takes copper longer to kill fungi than bacteria. 

8.  L. Weaver, H. T. Michels and C.W. Keevil, Potential for preventing spread of fungi in air-conditioning systems constructed using copper instead of aluminum. Letters in Applied Microbiology 50, pp. 18-23, 2010.


Viruses are organisms that can only reproduce inside the cells of a host organism, such as a bacteria, plant, or human cell.  Their genetic information is either DNA or RNA which is tightly compacted inside a capsule (capsid) envelope consisting of a number of proteins and lipids often arranged in very elaborate structures.  The purpose of the capsid is to facilitate attachment to the host cell, passage through the cell's outer layers/membranes, and thereby gain entry into the cell's cytoplasm, where all cellular production occurs.  Viruses can be permanently inactivated by destroying capsid function and/or morphology and thereby their ability to be infectious.

Influenza A Virus

Influenza A, a viral pathogen, causes hight rates of mortality and morbidity, especially in the elderly, and can survive on a range of environmental surfaces, including stainless steel. It is transmitted from person to person and to environmental surface  by viral-laden droplets, by hand to hand and hand to surface contact.  A short experiment was conducted to determine if copper can inactivate Influenza A.  Coupons of stainless steel and copper were inoculated with viral suspension and exposed at ~22 deg C and 50 to 60% RH and examined over time.  The results show below, where green dots indicate active virus particles as seen by epifluorescent microscopy, illustrate a reduction in virus particles from 2,000,000 units at time zero, to 1,000,000 at one hour and to 500,000 at six hours on the stainless steel.  The reduction on the copper is much faster, going from 2,000,000 million units at time zero, to 500,000 at one hour, to less than, or< 500 at six hours.  This illustrates that copper inactivates Influenza, and thus has the potential to reduce infections.

  9. L. Weaver, H. T. Michels and C.W. Keevil, Potential for preventing spread of fungi in air-conditioning systems constructed using copper               instead of aluminum, Letters in Applied Microbiology 50, pp. 18-23, 2010 


Infectivity of the pathogenic human coronavirus 229E following exposure to copper alloy surfaces was tested (Wilkins et al. 2015).  Coronavirus remained infectious following at least 5 days on a range of common surface materials, including Teflon, polyvinylchloride (PVC), ceramic tiles, glass, silicone rubber, and stainless steel.  Human coronavirus 229E was rapidly inactivated on a range of copper alloys.  The results are shown in the figure below, which was taken from the original article.  In Panels A and B, a sample containing 1,000 viral particles was dried onto six different copper/zinc alloys were tested ranging from 60% to 100% copper and pure zinc.  Stainless steel was the experimental control.  Panels C shows the results for 5 different copper/nickel alloys ranging from 70% to 100% copper, pure nickel, and the stainless steel control.  Taken together, panels A, B, and C clearly demonstrate that the copper in the alloy is the effect agent causing inactivation.  Moreover, hen only 50 viral particles were applied to the copper alloy samples to more realistically mimic hospital situations, inactivation was about 8-fold faster.  In summary, exposure to copper alloy surfaces rapidly inactivates coronavirus 229E and the rate of inactivation is faster in alloys containing higher concentrations of copper.  Inactivation seems to occur via a mechanism similar to the Membrane Targeting hypothesis (see Copper Killing Mechanism).  Wilks et al. (2015) found that exposure to copper destroyed the viral genomes and irreversibly affected virus morphology, particularly causing disintegration of the envelope and dispersal of surface spikes. 


10.  S. L. Warnes, Z. R. Little, and C. W. Keevil, 2015. Human Coronavirus 229E Remains Infectious on Common Touch Surface Materials,  mBio 6(6):e01697-15.


Norovirus is very contagious, causing numerous cases of gastroenteritis worldwide, many resulting in death.  It is transmitted by touching contaminated surfaces, hand to hand contact, and consuming contaminated food.  Outbreaks are common in hospitals and schools, and especially on cruise ships.  In addition, infected individuals shed norovirus for up to three weeks after they are symptom free.  The only way to control an outbreak is to use bleach to disinfect surfaces.  Human norovirus cannot be cultured in the lab and this has inhibited research.  However, murine norovirus, MV-1, is a close surrogate that can be cultivated.  MV-1 survival was challenged by a series copper alloys. The upper graph, shown below, shows the results from a series of copper nickel alloys and pure nickel as well as stainless steel, which is the experimental control.  Neither stainless steel nor pure nickel displayed any efficacy. The copper nickel alloys inactivate MV-1 and show decreasing efficacy as the copper content of the alloy decreases.  The lower graph shows the results from a series of copper zinc alloys as well a pure zinc and stainless steel, the experimental control.  Again, the efficacy correlates copper content with higher copper content causing killing in a shorter time.  As expected, stainless steel had no significant effect and pure zinc was about as effective as the alloy with the lowest copper content, C280.  Note that the pure zinc and C280, which contains 60% copper and 40% zinc show a very similar response.  Thus, there appears to be appears to be no synergy between zinc and copper in these copper zinc alloys.  These results indicate that copper nickel alloys with at least 79% copper and copper zinc alloys with at least 70% copper would be most effective against this virus.

  11. S. L. Warnes, E N. Summersgill and C. W. Keevil, Inactivation of Murine Norovirus on a Range of Copper Alloy Surfaces Is                               Accompanied by Loss of Capsid Integrity, Applied and Environmental Microbiology, 81(3), 1085-1091, 2015