Deprox fails log 6 test even in tiny test chamber.


deprox box

Hygiene Solutions claim a single Deprox unit has the capacity to decontaminate rooms with a volume of 380m3, e.g. a 12 bed ward bay. A typical hospital side ward (single room with ensuite) has a volume of 60m3.

However, the chamber used by Hygiene Solutions Ltd  to test the Deprox is 1.5m x 1.5m x 2.8m. Total volume 6.3m3 , just 10% of the volume of a hospital single bedroom, and 1.7% of the maximum volume that Deprox is guaranteed to disinfect. It is barely larger than a telephone box.

Hygiene Solutions internal testing, published here for the first time reveals that the Deprox, in spite of being boosted with 50% more concentrated hydrogen peroxide solution than the standard “Deproxin” was incapable of a log 6 decontamination of even this tiny test chamber.

The Deprox was thoroughly tested over a period of months by David Sempere Aracil, a well qualified chemist. David placed Log 6 biological indicator discs (Apex Biological Indicator #HMV-091) in 8 different locations around the inside walls of the test chamber. The Deprox unit (“Trusted by leading hospitals around the world”) was sealed in the chamber, and the process was run. The log6 BIs were incubated – they were all alive.

David tried substituting Sanosil SO15 which at 7.5% H2O2 is 50% more concentrated than Deproxin. Now some of the BIs would be sterilised, sometimes. Over several weeks in late 2014, David did a series of 12 tests in the test chamber, all with 7.5% H2O2 rather than the 5% Deproxin. He tried turning the ΔRH up and down, but to no avail. In 5 of these tests, all 8 BIs remained viable. None of the tests sterilised more than 6 out of 8.


In summary then:

Deprox, running on a 5% H2O2 solution, is claimed to give a log6 decontamination of an entire 380m3 ward, including inside small crevices and complex equipment. In Hygiene Solutions’ own tests, the Deprox running on a 7.5% solution, and thus generating a 50% higher aerial H2O2 concentration than the standard process, completely failed to give a log6 decontamination of a 6.3m3 box in multiple tests.

Hygiene Solutions continued to promote and sell the Deprox with exactly the same claims, but in 2015, they turned the whole Deprox fleet down from ΔRH20 to ΔRH5. See

Fortunately, (or unfortunately for Hygiene Solutions) David’s notes of these tests survived.


Explanation of table.

This table is a summary of 15 tests done by David Sempere Aracil, assisted by Tautvydas Karitonas, over a period of months. Both are university graduates with extensive research experience, and David has a PhD in Chemistry. The tests were done with a standard production model Deprox machine, the purpose of the tests was to determine if the extremely low efficacy of the Deprox process could be rectified by increasing the concentration of the hydrogen peroxide solution from the standard 5% to 7.5%.

The results were recorded in 3 A4 hardcover notebooks. Each of the 15 tests was recorded in more detail on preceding pages of the notebook. In addition to the efficacy tests, the notebooks contain extensive details of tests on prototype catalyst systems, and constitute proof that HS was well aware of both the low efficacy and residual gas issues with Deprox.

Heading: “Sanosil 015 forte” refers to Sanosil S015, which is a disinfectant intended for water systems. It is 7.5% Hydrogen peroxide solution. Note that this is more concentrated than the 5% Deproxin solution that is used in production Deprox machines.

Col.1. The test number. These are not sequential, as some tests did not use Biological indicators (BIs) and were not recorded in this resume.

Col. 2 Duration of test measured from when the machine starts vapourising. (It takes several minutes for the machine to fill the piezo tank at the beginning of each test)

Col. 3 Delta HR setting of machine. This is adjusted by using unmarked pressure sensitive switches below the LCD display. – see How to test your Deprox.

Col. 4 HRO This is the original relative humidity in the test chamber before the machine starts.

Col. 5 TO Temperature in the chamber before the machine starts

Col. 6 CMAX Hydrogen peroxide concentration in PPM, maximum level reached during process.

Results columns. The first 12 tests were done in the test chamber (wardrobe). Each number represents a specific marked location on the test chamber wall where an exposed stainless steel Bacillus subtilis log6 biological indicator was placed. The chamber is a crude wood and plasterboard structure in an essentially unheated warehouse. It is approximately 5’ x 5’ x 9’ and the indicators were placed at various heights on the interior walls of the chamber. The last 3 tests were done in the company board room which is approximately 12’ x 25’.

A” +” indicates that the BI still contained viable bacteria, a “–“ indicates that all bacteria on this indicator were killed.

Final column. This is the percentage of BIs that were killed.

Ultra-V, Ward 7B & the dark side of the moon.

Ultra-V moon

Spare a thought for the unfortunate patients of Yeovil Hospital wards 7B and 8B. With the two wards hit by a nasty Norovirus outbreak, Director of Nursing Shelagh Meldrum is proposing to eliminate the germs WITHOUT the use of bleach or other “harmful chemicals” (i.e. disinfectant). Instead, she will put her trust (and her patient’s lives) into the hands of the notorious fake medical equipment company, Hygiene Solutions Ltd of King’s Lynn.

Ultra-V yeovil hosp.

The Ultra-V sounds like something from science fiction. Unfortunately, that is exactly what it is. A cheaply reverse-engineered copy of an obsolete American device, built in the back of a farm sundries warehouse, the Ultra-V is foisted on unsuspecting NHS by unctuous salesmen armed with a pack of outrageous lies, which can be disproved by anyone with a basic understanding of physics.

UV-C light, as generated by the Ultra-V is well known for its germicidal properties. So what exactly is the problem with the Ultra-V, and why should the hospital not use this instead of bleach?

The problem is that light travels in straight lines, and hence casts shadows. This is very evident from the photo of a crescent moon above. The dark side is very dark indeed, the sunlight does NOT wrap around corners to light up the lunar night.  This is just as true for UV light as for any other wavelength. Otherwise you could get a suntan at night.

Shadowed areas in a hospital room are not exposed to the UV radiation, hence are not disinfected. Conventional UV-C systems deal with the problem by using two or three light emitting units placed around the room to eliminate the shadowed areas, or otherwise require the unit to perform two disinfection cycles from two different locations in the room, thus ensuring that all areas are exposed at least once.

According to their sales brochure, the Ultra-V, by some miracle of optical technology can reach “shadowed areas, under bedside units and hidden corners” all from “one central location within each room”.


This miracle is achieved by “Spectromes”. These are apparently small light meters that are placed around the room to ensure that all surfaces get fully exposed. The exposure time is theoretically extended until the darkest “Spectrome” has had a full dose.

The only UV light received by a Spectrome that is in a shadow is the diffuse reflections from the lit surfaces of the room. Unfortunately most substances absorb UV-C radiation very strongly – far more than they do for visible light. In the UV-C world, almost everything non-metallic looks black.

Typical hospital surfaces absorb 95% of UV-C radiation, and scatter the rest. It follows that the shadowed areas are very dark, as at the most they can expect to receive 1/20th of the radiation of the directly lit areas. Consequently, if this scattered light is to disinfect the shadows, the process will have to be extended in duration 20 fold. Is this what happens?


The normal process time with the Spectromes fully exposed is 15 to 20 minutes. Place a Spectrome in the shadows, and the process might extend to 40 minutes at the max. (Try it, if you don’t believe me.) At this point, the Ultra-V is programmed to override the Spectrome, and turn out the UV lights, indicating that the process is complete. Just in case hospital staff might be suspicious, the Ultra-V automatically sends a cheerful email to the operator’s designated address, giving time, date location, and certifying that the room has been decontaminated to a log 4 to 6 standard.

To give an example of how dangerous this is, consider the bedrails on a standard hospital bed – these are high touch areas constantly exposed to the patient’s hands and every cough and sneeze. From its “single central location” the Ultra-V unit, obviously and indisputably will only illuminate one side of each bed rail. The other side will be nearly as dark, in UV terms, as the dark side of the moon in the photo above. A standard Ultra-V process will leave these areas highly contaminated. A simple test with standard Biological Indicator coupons will prove this. (Again, don’t take my word for it – try it.)

To compound the error, the Spectromes are narrow bottomed plastic boxes that have to be placed on a hard, level surface, usually the floor or a bedside cabinet. it is completely impossible for them to monitor small, high touch areas like the back of a bedrail.

Spectrome Ultra-V

So on the positive side, at least the unshadowed areas will get a thorough log 4 to log 6 clean? Sadly not. The following Journal of Hospital Infection article detailing a study at the Wye Valley NHS Trust, demonstrated an efficacy of less than log 1 for the Ultra-V.


The outlook for the patients of Yeovil hospital does not look very bright, at any wavelength, unless there is a change of heart.

Please Shelagh Meldrum, for the sake of both patients and staff, don’t throw away the bleach just yet…use your £50,000 white elephant if you must, but give the wards a good old-fashioned hypochlorite deep clean first!


Fog shadowing – Deprox leaves 50% of surfaces untouched.

Rime 3

There has been much contention about the relative merits of vapour and fog based “HPV” systems, but the most important difference has been overlooked, which is that crude fogging systems such a Deprox will only reach on average 50% of the surfaces in a treated area. The effect has been masked, perhaps deliberately in some cases by the testing protocol, that always places BIs in the most exposed areas and favourable orientations.

In order to understand the problem, we need to shed some light on the difference between fogging and vapour based systems:

Definition of vapour

The scientific definition of a vapour is a gas, as opposed to an aerosol or fog. As a general rule, fogs and aerosols scatter light, and are visible as a white cloud. A gas or vapour is non-scattering, hence invisible.

How do vapour systems work?

Vapour phase systems emit invisible gaseous hydrogen peroxide into the air, and maintain the concentration at a high level (>100ppm) for about an hour, which will give a 6 log efficacy against most pathogens. The vapour is generated thermally from a 30% hydrogen peroxide solution.

How do fogging systems work?

Fogging systems emit an aerosol of small (5-10 micron) droplets of low concentration hydrogen peroxide solution (typically 5%). There are two different mechanisms of disinfection occurring simultaneously:

  1. The droplets impact on surfaces in the room and wet them with the solution.
  2. As droplets evaporate in the air, the hydrogen peroxide is released as a vapour, which will diffuse through the air and reach surfaces that are sheltered from droplet impact.

Vapour concentration is limited by Henry’s law to an average over the process time of about 50ppm. Hence these systems are capable of a log 4 efficacy, provided that the starting humidity of the room is low enough to allow the droplets to evaporate.


Hydrogen peroxide is a “lazy gas”. Its high molecular mass gives it a slow rate of diffusion, hence auxiliary fans should always be used to ensure the gas is thoroughly mixed and distributed throughout the room to be treated. If this is done, a vapour system will give a homogeneous distribution of gas, which will disinfect all surfaces regardless of orientation or distance from the generator. The H2O2 molecules break down continuously, with a half life of about 50 minutes, so if left to diffuse naturally, the concentration will drop substantially with distance from the machine.

Fog Shadowing

Fogging systems have altogether different dynamics. The fog droplets fall continuously under gravity relative to the surrounding air. For example, a 2 micron fog droplet will fall by 50 times its diameter per second. This has a dramatic effect on the distribution of the active ingredient on the surfaces in the room. In order to wet a surface, and thus transfer the H2O2 to the pathogens present, the droplets have to impinge on it for sufficient time and with sufficient force to break the surface tension. For upwards facing horizontal surfaces, the gravitational settlement is adequate, and these surfaces will have a visible film of moisture at the end of a process. A horizontal downwards facing surface, such as the underside of a table or door handle will remain dry.

Unless the airflow direction is varied, i.e. by multiple oscillating fans, droplet contact will still be extremely uneven, with almost all the drops impacting on the side of the object facing the airflow. This is perfectly illustrated by this picture of a rime frost. Here a combination of a supercooled fog and a light breeze has caused the droplets to impact and freeze on the side of the fence wire facing the wind. The opposite side of each wire has no ice at all. This is very similar to the shadowing effect of the Ultra-V systems – on average, 50% of surfaces are “shadowed” and hence untreated.


Disinfection of walls and other vertical surfaces is a lottery. The degree of wetting, and thus disinfection is at the whim of the air currents in the room, and droplets impinging on the wall at a shallow angle will bounce, particularly if the surface is at all hydrophobic.

None of this would matter if the vapour level was adequate, as the vapour would disinfect the surfaces not wetted by the droplets. However, in humid conditions, the vapour level is greatly reduced – leading to a very patchy and inadequate performance.

Temperature and circulation

The emission from fogging systems is cold – as much as 15 degrees cooler than the room temperature, due to the cooling effect of droplet evaporation. This cold dense vapour falls rapidly to floor level. The effect can be very visible on starting machines that eject the vapour vertically – the cloud of fog will often not reach the ceiling before collapsing and flowing down to the floor. In the case of the Deprox, the air inlet is on the bottom of the machine, so creating a circulation cell where the fog rises through the machine in the centre of the room and in the rest of the area is moving downwards.
Deprox convectionThe effect of this is to give heavy droplet wetting on the upwards facing surfaces around the machine. Bed rails, for example will be wet on the top and dry on the bottom. This has the fortuitous (for the manufacturer) effect, that if the machine is tested in the customary manner by placing biological indicators around the room in petri dishes facing upwards these samples will be exposed to a substantial “rain” of droplets and may show high efficacy levels which are not at all representative of the real disinfection achieved. If the BIs were secured in a vertical or downwards facing orientation, much lower efficacies would be recorded.

Perhaps the best way to determine the real efficacy of a fogging system in a typical side ward application is to attach the BIs to the walls of the ensuite. This is usually the area most distant from the machine, and also the most contaminated. Attaching the BIs to the walls in a vertical orientation will remove the effects of falling droplets.

Effect of humidity

Because both types of system are evaporating an aqueous solution, the humidity in the room rises during the process as water is evaporated into the air along with the H2O2.

The more concentrated 30% solution used by true vapour systems obviously means that less water is evaporated, and these systems can achieve the target aerial concentrations even in conditions of high initial humidity.

By contrast, the fogging systems using a 5% solution must evaporate 19 units of water for  each unit of H2O2. They cannot function at all over about 70% starting humidity, and in anything but the driest weather conditions, the level of disinfection achieved will be limited. This is because the evaporation of the initial outflow of droplets quickly raises the humidity to saturation, and fog emitted later in the cycle cannot evaporate, remaining as a highly visible white cloud.

Most fogging systems claim to be a “dry mist” and are set to avoid condensation, as this can damage electronics and stain furnishings. The only way to avoid condensation is to turn the machine off when the humidity reaches 90% or so. Thus while the Deprox has a fixed cycle time,  in humid weather, firstly much of the H2O2 is locked up in the fog droplets, and secondly much less solution used by the system, as it will spend most of the cycle “waiting” for the humidity to drop to a level where its control system will restart the fog generation. (Remember that the H2O2 is continually breaking down, so without continuous replacement, the concentration drops rapidly.)

Even this automatic regulation seems to have been inadequate, as former employees of Hygiene Solutions report that they were told to turn down the units to very low levels at times of high humidity to avoid problems.

This is not a minor effect – Deprox users will have noticed how on some days almost no fog is visible, while on others there is a dense fog which is still visible at the end of the “deactivation” cycle. It is also unsurprising that it is the systems in south Wales that have been suspended – the high humidity in this area leads to frequent process fails and residual fog.



There are now fogging systems on the market that use electrostatics to give an even distribution of droplet wetting regardless of orientation. This is a technology that has been used in paint and agricultural chemical spraying for decades, and causes the fog to “wrap” around objects and coat all faces evenly. I have no data as to the efficacy of this process, but it is a promising idea.

More detailed data about the differences between fogging and vapour systems can be found at