Originally Published MDDI May 2001
STERILIZATION
Exploring the Feasibility of Using Dense-Phase Carbon Dioxide for Sterilization
Though still in the research stages, sterilizing with CO2 may soon be a viable optionboth economically and logisticallyfor medical device manufacturers.
Michael A. Matthews, Langdon S. Warner, and Heinz Kaiser
![]() |
| A basic schematic of the proposed CO2 sterilization chamber. |
The conventional sterilization
processes commonly used by medical device manufacturersEtO, steam, gas plasma,
and gamma irradiationare proven technologies. By gauging the requirements
of their products and doing cost comparisons for various methods, manufacturers
settle on a sterilization process of choice. But a new sterilization process for
medical devices, dense-phase carbon dioxide (CO2), may change some
manufacturers' opinions. Although still in the research phase, sterilizing medical
devices using dense-phase CO2 may prove to be a viable alternative
both for OEMs and for end-users at healthcare facilities.
This article explores the technical and economic feasibility of a possible
dense-phase carbon dioxide (CO2) sterilization process, comparing
it to currently available commercial processes. Dense-phase CO2 has
been shown to deactivate some microorganisms. Although research is ongoing,
CO2 may prove to have certain technological, environmental, and safety
advantages compared with existing sterilization methods.
For the purposes of this article, the costs of sterilization using available
equipment were determined from a survey of operations in a major research facility.
The capital and operating costs of the hypothetical CO2 sterilizer
were estimated based on similar equipment that is currently being developed
for other applications.
CO2BACKGROUND
The Supercritical Fluid State. When gaseous or liquid CO2
is heated and compressed above the critical temperature (31°C) and pressure
(73 atm), it becomes a dense, highly compressible fluid that demonstrates properties
of both liquid and gas. Substances in the supercritical fluid (SCF) state normally
have better solvating ability than do the same substances in the liquid state.
The SCF has a viscosity similar to that of a gas, but a density closer to that
of a liquid. The properties of an SCF can be changed by adjustments of pressure
and temperature. Because of the low viscosity of SCFs (compared with familiar
liquid solvents), an SCF can penetrate small orifices. Additionally, substances
disperse throughout the SCF rapidly, due to high diffusion coefficients.
1
Dense-Phase Fluids. The liquid phase of CO2 requires temperatures
to be below ambient, but pressure can typically be decreased to between 700
and 1000 psi. The lower pressure decreases costs, but some of the unique properties
of the SCF state are sacrificed as a result of the lower pressure. Nevertheless,
the liquid state of CO2 conserves some of the solvent properties
of the supercritical statenamely low viscosities, high diffusion coefficients,
and acceptable solvent power. Collectively, the term dense-phase fluid
refers to operations in either the supercritical or liquid states.
Environmental Benefits. CO2 is of particular interest in
dense-fluid technology because it is inexpensive, nonflammable, nontoxic, odorless,
and nonpolluting. Manufacturersor healthcare workers and patients at the
end-user stagewould not be exposed to toxic residues using CO2-based
sterilization. Some current sterilization and disinfection methodsEtO
processes, for exampleuse toxic and flammable substances that have a strong
odor.
CO2 is not a hazardous substance regulated by the Environmental
Protection Agency, and its use in sterilization would not contribute to the
greenhouse effect because CO2 would be captured as a by-product from
other industrial activity.
COMMON STERILIZATION METHODS
Steam. Steam sterilization is widely used by medical device manufacturers
and end-user facilities because it is effective, efficient, safe, and inexpensive.
It cannot be used to treat heat-sensitive materials, however. The main effect
of steam is to denature the proteins in the microorganisms and break down the
microbial proteins, lipids, carbohydrates, and nucleic acids.
EtO. EtO is a colorless, flammable, and highly reactive compound. Due
to its low boiling point of 10.4ºC (50.7ºF) at atmospheric pressure,
EtO behaves as a gas at room temperature. EtO chemically reacts with amino acids,
proteins, and DNA to prevent microbial reproduction.
2,3 This alkylating agent is widely used to sterilize
heat-, radiation-, and moisture-sensitive materials such as those containing plastics
and microelectronics. It should not be forgotten, however, that EtO is toxic and
explosive. EtO sterilization facilities require extra ventilation and safety precautions.
Gas Plasma. Like EtO, hydrogen peroxide plasma is used for the sterilization
of heat-sensitive devices that cannot be treated above 60ºC. Plasmas can
be created at high or low temperatures in strong electromagnetic fields. The
advantage of using plasmas is the absence of toxicity after the electromagnetic
field is shut down. 4
In this method, sterilization is achieved by the use of hydrogen peroxide gas
plasma for 45 to 75 minutes total cycle time, at low temperatures and in an
environment of low moisture.5
| Method | Sterilizer | Number | Loads February | Loads March |
| Steam | Amsco 3053 | 5 | 499 | 606 |
| Ethylene Oxide | Amsco 3017 | 5 | 82 | 93 |
| Hydrogen Peroxide | Sterrad 100 | 1 | 15 | 73 |
Table I: Loads processed in the study facility's sterile processing department during February and March.
SURVEY OF STERILIZATION OPERATIONS
To provide a quantitative basis for comparing CO2 sterilization to existing technology, the sterilization operations at a major South Carolina hospital facility were surveyed. At this location, there are 48 machines performing sterilization that are dispersed among the different departments in the facility. The predominant sterilization method is steam, selected because of its cost advantages and effectiveness. The sterile processing department, however, has EtO and hydrogen peroxide gas plasma sterilizers in addition to the steam systems. These latter two methods of sterilization are used with heat-sensitive instruments.
Due to the diversity of the machines, chamber sizes, cycle times, rates of usage, and methods of sterilization, a cost per cubic foot was determined in order to compare costs among the different sterilization methods. The cost per cubic foot was determined by dividing the total monthly cost by the number of cubic feet processed in a month. The total monthly cost consists of the mortgage payment on the capital invested in the machines (with a fixed 7.5% interest rate for 10 years), the monthly portion of the respective maintenance cost, and the operational cost of processing each load in a given month.
The sterile processing department is the only department in the facility that sterilizes items with methods other than heat and steam. As a result, this department was more closely studied. Information about the number of loads was collected for the months of February and March 1999. Table I shows the number of loads during these two months and identifies which sterilization methods were used.
The steam loads were processed in Amsco 3053 machines, each with a capacity of 30 cubic feet. The EtO loads were sterilized with Amsco 3017 sterilizers, each with a chamber size of 5 cubic feet. The hydrogen peroxide gas plasma loads were processed in a Sterrad 100 with a 3.5-cubic-foot chamber. The cost comparison for the different sterilization methods in the sterile processing department is shown in Figure 1.
![]() |
| Figure 1. Cost comparison for the study's sterilization methods. |
Personnel. The personnel cost for the sterile processing department was surveyed in order to obtain the personnel cost per cubic foot of sterilizing surgical instruments. At the time of the survey, there were three shifts. The function of these employees is to sort, disassemble, clean, assemble, inspect, pack, and wrap the items to be sterilized. This activity generates an additional dollar amount per cubic foot required for each of the different sterilization methods.
The personnel cost for processing heat-resistant devices is approximately $7 per cubic foot in the sterile processing department. Heat-sensitive devices have a higher personnel cost per cubic foot, however, at $37.
THE PROTOTYPE CO2 STERILIZATION PROCESS
Based on several patents,
612 as well as the experiments performed
by several researchers,1320 a hypothetical
set of process conditions was chosen for a prototype CO2
sterilizer. The primary consideration is pressure, which may be as high as 200
atm or more in industrial SCF processes. The temperature required for supercritical
CO2 sterilization is close to room temperature, ranging from 35º
to 45ºC. Agitation of the CO2 is an important factor that may
reduce the time required for sterilization. The water content may also affect
the sterilization cycle time. Higher water content improves the permeability of
an organism, allowing the CO2 to penetrate the cell wall and kill it.
Although the method of sterilization using supercritical CO2
has not to date been completely defined by research, the general procedures are
similar among the different researchers. This prototype CO2 sterilization
cycle uses the following steps:
CAPITAL AND OPERATING COSTS
To determine the capital and operating costs of a hypothetical dense-phase
CO
2 sterilizer, several manufacturers of existing
commercial equipment were contacted, along with several experts in the field.
Commercial applications include dry cleaning and parts cleaning with CO2.
These processes operate at conditions similar to those expected for the CO2
sterilization process. The annual maintenance cost, according to equipment manufacturers,
ranged from 3 to 10% of the cost of the equipment.21
The operational cost per load is determined by considering technical aspects
of a CO
2 machine and the supplies required to comply
with terminal sterilization (sterilization of sealed containers) standards. The
variables affecting the cost per load are size of the chamber, degree of CO2
recycling, cost of CO2, cost of power, cycle
time, and cost of supplies.
The Chamber. The chamber size determines the amount of CO
2 needed to sterilize or clean the load. In this
study, the chamber size is 3.5 cu ft. The amounts of CO2
required for a 3.5-cu-ft chamber in the liquid and supercritical stages are 180
and 205 lb, respectively. Recycling minimizes CO2
losses, but also increases the power consumption. The recycling percentage selected
for this analysis is 95%.
Although the cost of CO2 is relatively low
($0.20 to $0.25 per lb), it is an important factor in the cost per load. The
CO2 cost used in this analysis is $0.20 per pound. The hourly operating
cost for this analysis is $1.50.
Supplies. There can be no accurate information given about the cost
of supplies for CO2 sterilization because these supplies are specific
for each sterilization method. For example, the pouches and wraps used in the
hydrogen peroxide gas plasma machine are different from those used in EtO and
steam sterilization.
Wraps and pouches allow the sterilant to come in contact with the items being
sterilized. Other supplies that must withstand the sterilization process include
biological indicators, chemical indicators, and tape. For the purpose of this
article, it is assumed that the cost of these supplies is similar to that considered
in the EtO and gas plasma hydrogen peroxide methods. The cost per load of these
supplies is approximately $8.25.
The cycle time influences the number of cubic feet that a sterilizer is able
to process. Shorter sterilization cycles improve the turnover rate of the materials,
offering more flexibility and decreasing the investment in additional equipment.
A cycle time of 2 hours was selected for the hypothetical CO2 sterilizer.
The cost per cubic foot for CO
2 sterilization was determined following the
previous model developed for steam, EtO, and hydrogen peroxide. The capital, maintenance,
and operational costs were added, and this sum was divided by the maximum number
of cubic feet that the CO2 sterilizer would process
per year. The principal variables influencing the cost per cubic foot of CO2
sterilization, and the cost per cubic foot, are shown in Table II. This table
depicts the conditions and values of probable equipment.
Table
II: Variables affecting the cost of CO2 sterilization
(prototype). STERILIZATION METHOD COMPARISON
The steam sterilizer chosen as a basis for this study is an Amsco 3053-S prevacuum
currently used in the facility's sterile processing department. This sterilizer
has a maximum capacity of 30 cu ft for the sterilization of porous-, heat-,
and moisture-stable goods. A microcomputer system monitors and controls all
the operations and functions during the sterilization process.
The EtO sterilizer used for comparison is an Amsco 3017 that uses 100% EtO.
This sterilizer, with a capacity of 5 cu ft, is equipped with an aerator to
reduce the EtO levels in the breathing zone. A single-dose gas cartridge provides
the 100% EtO, which is enough gas sterilant to process one load. This system
also has a microcomputer that monitors and controls operational variables
The Sterrad 100 is a hydrogen peroxide gas plasma sterilizer produced by Johnson
& Johnson. This sterilizer is able to process loads up to 3.5 cu ft. The
installation and operating requirements consist of a simple connection to electricity;
no water, steam, or air supply is needed. All of the operations are controlled
and monitored by a microcomputer, which generates system performance records
for each cycle. A cassette with 10 individual doses provides the hydrogen peroxide.
In the case of a prototype CO
2 sterilizer, it is assumed that a 3.5-cu-ft
vessel is used. The vessel is constructed of stainless steel in order to support
the elevated pressures needed to reach the supercritical stage (200 atm). The
method of agitation is important, but the precise details must be determined through
further research. CO2 machines must be designed with microcomputers
that control and monitor the pressure, temperature, and time. For quality assurance
and safety reasons, they must also register and print sterilization time, fluid
conditions, and other parameters as is done with steam, EtO, and hydrogen peroxide
sterilizers.
Hydrogen peroxide and CO2 sterilizers operate without process steam,
air, or water. Moreover, they do not require drain or gas exhaust to dispose
of their process waste. The two-hour CO2 cycle
time is likely to compete with the Sterrad cycle of 75 minutes and is much shorter
than the 15-hour cycle required by the EtO sterilizer.
COST BENEFITS OF CO2
In order to introduce a new sterilization technology, of course, it must be
cost competitive. The technological advantages alone may not be enough to justify
replacing capital equipment if the costs do not benefit customers and manufacturers
alike. The capital costs of new technology are normally higher than those of
established technology due to the investment made and the lack of economies
of scale. Presumably, the estimated cost of the machine would decrease over
time as a result of competition among different manufacturers and additional
technological improvements.
Table III summarizes the cost comparison based on the assumptions and data
described above. The maintenance cost of CO
2 sterilization is similar to that of a Sterrad
100; however, the cost of maintenance does not strongly influence the cost per
cubic foot. The availability of service affects the decision of acquiring sterilizers.
Service is an important factor for current customers of steam, EtO, and hydrogen
peroxide gas plasma sterilizers. For example, both Steris and Johnson & Johnson
offer a coast-to-coast service for repair or preventive maintenance. Other competitors
offer similar service in the United States and abroad. Companies trying to commercialize
CO2 sterilization would necessarily have to offer
a similar service.
The cost per cubic foot for the different sterilization methods puts all these
sterilization technologies on the same basis. Capital, maintenance, and operational
cost are summed in a single cost. The cost per cubic foot for steam sterilization
($1 per cu ft) is the least expensive of all the methods considered in this
analysis. This is as expected due to the advantages of steam sterilization previously
mentioned. The cost per cubic foot of CO
2 sterilization is lower than that of EtO ($6
versus $19). Factors increasing the cost per cubic foot for EtO are the long cycle
time, risk management activities, insurance, legal liability, and monitoring equipment,
among others. Based on this cost analysis, CO2
and H2O2 sterilization
would most closely compete due to the cost and technical advantages.
Variable
Parameter
Cost
Capital
cost
Chamber
size: 3.5 cu ft
$170,000
Maintenance
5%
of capital cost
$8300/year
Operational
cost
Recycling
95% of 181 lb $0.20 per lb
$1.80
per run
Cycle
time 2 hours at $1.50 per hour
$3
per run
Pouches,
wraps, BI, CI, tape
$8.25
per run
Total
cost per run
$13
Cost
per cubic foot
$5.9
| Cost | Steam | EtO | H2O2 | CO2 |
| Machine | 70,000 | 40,000 | 123,000 | 170,000 |
| Maintenance | 1100 | 2500 | 8300 | 8300 |
| Cost per load | 25 | 77 | 17 | 13 |
| Cost per cu ft | 1 | 19 | 6 | 6 |
Table III: Cost comparison, in dollars, of the four sterilization methods.
CONCLUSION
CO2 sterilization may prove to be a viable substitute for current
EtO sterilization of heat-sensitive materials and devices. This analysis of
CO
2 sterilization shows a lower cost per cubic
foot ($6) than EtO ($19) because of the shorter cycle time, lower cost per load,
and lack of regulatory constraints. Moreover, CO2
does not have the potentially negative environmental and health effects of EtO.
Hydrogen peroxide gas plasma sterilizers have a shorter cycle time than prototype
CO
2 sterilizers. Their cost per cubic foot is the
same ($6), and both technologies have no impact on the environment or on manufacturing
employees.
The present analysis is, of course, preliminary. Further research and development
is required. CO
2 will have disadvantages. For example, the capital
cost may be significantly higher than it is with existing technologies, even though
the cost per cubic foot is lower than with EtO and equal to H2O2.
Another practical problem would be the space needed to store the CO2
cylinders. Even though most of the gas is recycled, CO2
sterilization requires an inventory of several hundred pounds of CO2
per cycle. This could be a prohibitive constraint when comparing sterilization
technologies.
CO
2-based technology is technically feasible for
sterilization within the parameters suggested by available research. Equipment
manufacturers from other applications (e.g., dry cleaning and precision cleaning)
should be able to design and build a prototype that would satisfy the temperature,
pressure, humidity, and agitation requirements for CO2
sterilization. CO2 is widely available and relatively
inexpensive. Several companies distribute and sell CO2
in different grades.
The CO
2-based technology may be feasible and profitable
if the application (market) is carefully targeted. The most likely market for
CO2 among existing applications is as a replacement
for EtO sterilizers. As noted, CO2 sterilization
is less expensive than EtO sterilization. In addition, it does not pose the kind
of environmental and health risks associated with the use of EtO. Hydrogen peroxide
gas plasma sterilization is competitive with CO2
sterilization in that the two technologies have similar costs, their cycle times
are approximately the same, and they do not produce toxic wastes.
Steam sterilization is the most effective method for the sterilization of
heat-resistant instruments, and the costs of autoclaving are lower than those
of CO
2 sterilization. It is not proposed that CO2
competes with steam for the sterilization of heat- and moisture-stable goods.
But CO2 sterilization is a suitable method for
the sterilization of heat-sensitive devices. The current trend in hospitals is
to increase the number of less-invasive medical procedures, which are performed
with complex devices that require low-temperature sterilization methods.
Finally, though the use of CO
2 sterilization offers cost and environmental
advantages, there is no guarantee that a CO2
sterilizer will receive FDA approval. A sterilizer must achieve the sterility
assurance level under the worst conditions, and it must be capable of meeting
the sterilization specifications every time the process is performed. The process
must maintain the functionality of the product and its packaging. All of these
areas require further research.
REFERENCES
1.AD Russell, WB Hugo, and GAJ Ayliffe, Principles and Practice
of Disinfection, Preservation and Sterilization (Blackwell Science, Oxford,
UK: 1999).
2.PJ Landrigan, TJ Meinhardt, and J Gordon, "Ethylene Oxide: An Overview of
Toxicologic and Epidemiologic Research," JAMA, 255 (1986) 1575.
3.DJ Hurrel, "Recent Developments in Sterilization Technology," Medical
Plastics and Biomaterials, 5 no. 5 (1998) 2637.
4.PT Jacobs Advances in Low-temperature Gas Plasma Sterilization Technology.
Johnson & Johnson (1999).
5.G Park and A Crull, Supercritical Fluids: Their Time Has Come. (Business
Communications Co., Norwalk, CT: 1994).
6.A Dillow, F Dehghani, J Hrkach, N Foster, and R Langer, Supercritical
Fluid Sterilization Method. (PCT/US99/14266, Dec. 29, 1999).
7.Y Osajima, M Shimoda, T Kawano, and K Okubo, "Method for Inactivating Enzymes,
Microorganisms, and Spores in a Liquid Foodstuff," U.S. Patent 5,667,835, Sep.
16, 1997.
8.Y Osajima, M Shimoda, T Kawano, and K Okubo, "Method for Modifying the Quality
of Liquid Foodstuff," U.S. Patent 5,520,943, May 28 1996.
9.Y Osajima, M Shimoda, T Kawano, and K Okubo, "System for Processing Liquid
Foodstuff or Liquid Medicine with a Supercritical Fluid of Carbon Dioxide,"
U.S. Patent 5,869,123, Feb. 9, 1999.
10.Y Osajima, M Shimoda, T Kawano, and K Okubo, "System for Processing Liquid
Foodstuff or Liquid Medicine with a Supercritical Fluid of Carbon Dioxide,"
U.S. Patent 5,704,276, Jan. 6, 1998.
11.P Drumheller, "Materials and Methods for the Immobilization of Bioactive
Species onto Polymeric Substrates," U.S. Patent 5,914,182, Jun. 22, 1999.
12.D Jackson, S Barcotta, and M Lepp, "Process for Cleaning, Sterilizing,
and Implanting Materials Using High Energy Dense Fluids," U.S. Patent 5,213,619,
May 25, 1993.
13.A Dillow, F Dehghani, J Hrkach, N Foster, and R Langer. "Bacterial Inactivation
by Using Near- and Supercritical Carbon Dioxide," Rp. Natl. Acad. Sci. USA
96 no. 18 (Aug. 31, 1999) 1034410348.
14.A Enomoto, K Nakamura, K Nagai, T Hashinoto, and M Hakoda, "Inactivation
of Food Microorganisms by High-pressure Carbon Dioxide Treatment with or without
Explosive Decompression," Biosci. Biotech. 61 no. 7 (1997) 11331137.
15.H Ishikawa, M Shimoda, H Shiratsuchi, and Y Osajima, "Sterilization of
Microorganisms by the Supercritical Carbon Dioxide Micro-Bubble Method," Biosci.
Biotech. Biochem. 59 no. 10 (1995) 19491950.
16.M Kamihira, M Taniguchi, and T Kobayashi, "Sterilization of Microorganisms
with Supercritical Carbon Dioxide," Agric. Biol. Chem. 51 no. 2 (1987)
407412.
17.H Kumagai, C Hata, and K Nakamura, "COCLASS=s10>2 Sorption by
Microbial Cells and Sterilization by High-pressure CO2." Biosci.
Biotech. Biochem. 61 no. 6 (1997) 931935.
18.K Nakamura, A Enomoto, H Fukushima, K Nagai, and M Hakoda, "Disruption
of Microbial Cells by the Flash Discharge of High-pressure Carbon Dioxide,"
Biosci. Biotech. Biochem., 58 no. 7 (1994) 12971301.
19.Taniguchi, M., Suzuki, H., Sako, M., Kobayashi, T. "Sterilization of Plasma
Powder by Treatment with Supercritical Carbon Dioxide," Agric. Biol. Chem.,
51, (12), 34253426 (1987).
20.CI Wei, MO Balaban, SY Fernando, and AJ Peplow, "Bacterial Effect of High
Pressure CO2 Treatment on Foods Spiked with Listeria or
Salmonella," Journal of Food Protection 54 no. 3 (1991) 89193.
21.H Kaiser, "Technical and Market Feasibility of Using Dense Phase Carbon
Dioxide for Sterilization in Hospitals," Master's Thesis, University of South
Carolina, December 2000.
Michael A. Matthews,
PhD, is a professor of chemical engineering at the University of South Carolina
(Columbia, SC). He conducts research on dense-phase carbon dioxide. Langdon
Warner, PhD, is a research professor and a consultant on pollution prevention
and waste minimization. Heinz Kaiser is an industrial engineer who obtained
his master's degree in earth and environment resources management at the University
of South Carolina. Copyright ©2001 Medical Device & Diagnostic Industry





