Application Notes and
White Papers - Particle Contamination
Particle Counting and the FDA
By Joe Gecsey, Product Manager, Pacific Scientific
This article explores the FDA's approach to
monitoring particle contamination and the need to protect particle counters in
humid or washdown environments. The article was compiled frorn conversations
with the FDA and operators of sterile fill lines.
THE IMPACT OF VARIOUS government regulations on actual manufacturing processes
are not always obvious. The current government requirements for monitoring
sterile fin lines in cleanrooms requires a specific method of particle counting
that is somewhat different than Federal Standard 209D. 'Me particle counting
system will have to be compatible with the manufacturing environment, sample the
correct locations at the right size. and fulfill the FDA's requirement for batch
The FDA currently requires the manufacturer to show that the process area was in
control only during the actual filling operation. Continuous monitoring of the
whole filling process is preferred, but not required. Monitoring Is not required
when the room is not in use. It is good practice to monitor during set-up to be
assured of the quality of the air, but it is not required.
If the product or personnel are changed during a shift, a new monitoring program
must be started. The particle counts taken during the morning do not reflect the
operating conditions later in the day if a new product or new personnel have
been introduced. Changing shifts would also require a new verification of the
mom, even if the same product was being made. The monitoring procedure must
occur during the actual filling operation, with the equipment and personnel
operating in their typical modes. This approach makes good sense in a majority
of cleanroom operations.
The monitoring of microorganisms in the product and in the environment is still
the main focus for assuring product sterility: monitoring of particles - living
or not-is useful as an additional indicator of control and as a "real-time" indicator of potential contamination.
Federal Standard 209D
Although the FDA uses Federal Standard 209D as a general basis for their
guidelines. the FS-209D Is not an absolute set of rules for the pharmaceutical
industry. One of the essential differences between the FDA's philosophy and
FS-209D is that FS-209D allows averaging of counts across sample points. During
batch filling, averaging data across multiple sampling points is not acceptable
to the FDA. To keep proper records for a sterile fill area. the particle counts
obtained at each critical point must be maintained as a separate record. The
intent here is to monitor the air approaching the filling line, not to
characterize the general particle level of the room.
The "Guideline on Sterile Drug Products produced by Aseptic
Processing" states that the sample points should be within one foot of the
fill line. The FDA will accept a greater distance if It can be shown that
sampling within one foot will produce erroneous counts due to some processing
condition. such as over spray from a highspeed filling needle or a powder
Also In contrast to FS-209D. the "Guideline ... for Aseptic
Monitoring" requires monitoring at 0. 5 microns: monitoring at other size
ranges, for example 1.0 or 0.3 microns, must be well justified for the
particular filling process. For aseptic areas. the counts at each sample point
must be below 100 counts per cubic foot for all particles 0.5 microns and
larger. Ibis generally equates to an FS 209D "Class 100." but the FDA
specifically requires monitoring at "0.5 micron and larger" sizing.
Using alternate size equivalents is not acceptable unless the manufacturer can
justify an alternate size based on unique processing requirements.
Establishing Quality and Control of the Processing Area Using Trends
At a given sample point, repeated monitoring will provide a typical "baseline" of values during the manufacturing operation.
"Alert" and "Action" levels can be set relative to the
empirically determined baseline count. Using 1 -sigma and 3-sigma points derived
from the baseline count might be appropriate. The FDA is comfortable with
averaging data at a given sample site: that is. each and every sample does not
have to be below the "Alert" level to continue the processing the
batch. However, a consistent trend away from the baseline should be a signal
that the area is out of control.
Selection of Sample Points
Within the sterile fill area. samples must be taken in the immediate vicinity of
the fill heads, and preferably near where the containers enter the sterile area.
In general. you should monitor at points where the product Is exposed to
potential contamination and where an operator might frequently be present.
Outside of the sterile fill area or other "critical" areas. the FDA
recommends frequent monitoring to ensure that the HVAC filtration system is in
order, but the frequency of monitoring can be relaxed to between daily and
once-per-week. Again. the intent of gathering this data is to show that the area
is generally in control.
Short extensions - up to a maximum of roughly 10 feet - from the sample point to
the sensor are generally acceptable, assuming that the tubing has a minimum of
turns or curves and that the curves have a generous radius. Due to the
statistically low number of particles within a sample under "Class
100" conditions, it is best to limit the use of tubing, which causes some
entrapment or fragmentation of particles. If the tubing must be longer than 10
feet. then the loss factor for that given tubing must be determined and a
correction factor must be used to adjust the counts obtained during filling
procedures. In general, the use of manifolds for sampling in extremely clean
areas (e.g. Class 10) Is strongly discouraged by the FDA.
In general. for most sterile products. the particle count data should be saved
for a period extending one year past the expiration date of the product. There
is no restriction on the form of the data storage: it must simply be retrievable
for inspection within a reasonable time.
Sterile fill areas are often aseptically washed. However, particle counters must
be protected from moisture. Since tubing length restrictions require the
particle counter/sensor to be in the washed area, the equipment must be housed
in a moisture and splashproof container. such as a NEMA 4X rated enclosure.
Summary of ASHRAE Standard 52.2P
By Paul F. Johnstone
3. Definitions and Acronyms
Subject- A.S.H.R.A.E. Standard 52.2P Purpose: To summarize the standard where it
applies to particle counting and define the role of Particle Counters in 'this
application. General: Prior to 1996 the ASHRAE standard (52.1 -1992) utilized a
gravimetric method of measuring airborne particulate which weighed accumulated
particles, expressed in micrograms per cubic meter. In 1996 'the standard was
amended to include 'the use of Particle Counters as a method of particle count
and size measurement.
Section 1) Description and Summary of the Standard
This standard addresses air cleaner performance characteristics of importance to
the users; the ability of a device to remove particles -from the airstream and
it's resistance to airflow. Air cleaner testing is conducted at airflow rates of
not less than 510 cfm nor greater than 3180 cfm. A sample of air from a general
ventilation system contains particles with a broad range of sizes having varied
effects, sometimes dependent on particle size. Coarse particles for example,
cause energy waste when they cover heat transfer surfaces. Fine particles cause
soiling and discoloration of interior surfaces and furnishings as well as
possible health effects when inhaled by occupants of the space. When air
cleaners are tested and rated for efficiency in accordance with this standard,
there is a basis for comparison and selection for specific tasks. The test
procedure uses laboratory generated potassium chloride particles dispersed into
the airstream as the test aerosol. "A particle counter measures and counts
the particles in 12 size ranges both upstream and downstream for the efficiency
This standard establishes a test procedure for
evaluating the performance of air cleaning devices as a function of particle
2.2 The standard defines for generating the aerosols required for conducting the
test. "The standard also provides a method for counting airborne particles
of 0.3µ to 10.0µ in diameter upstream and downstream of the air cleaning
device in order to calculate removal, efficiency by particle size."
2.3 This standard also establishes performance specifications for the equipment
required to conduct the tests, defines the methods of calculating and reporting
the results obtained from the test data and establishes a performance
classification system which can be applied to air cleaning devices covered by
3.1 Definitions Correlation Ratio, R: The ratio downstream to upstream particle
counts determined from "the average of at least three samples taken during
a "no-filter" or "0% efficiency" test". This ratio is
used to correct for any bias between upstream and downstream sampling systems
and it's derivation is described in 10.3.
Isokinetic Sampling Is that in which in the sampler inlet is moving at the same
velocity and "direction" as the flow being sampled.
Test Aerosol: Poly disperse solid-phase (i.e., dry) Potassium Chloride (KCL)
particles generated from an aqueous solution, used- in this standard to
determine the particle size efficiency of the device under test. Generation of
the aerosol is described in 5.3.
Device: Throughout 'this standard the word "device" means air cleaning
equipment such as a filter.
3.2 Acronyms ASHRAE: American Society of Heating, Refrigerating, and Air
Conditioning Engineers, Inc. CV: Coefficient of Variation HEPA: High efficiency
particulate air PSE: Particle size removal efficiency ULPA: Ultra-low
penetration air 4 Test Apparatus
4.2.2 Room air shall be used as the test air source. The temperature of the air
at the test device shall I be between 50 and 100°F with a relative humidity of
between 20% and 65%.
4.3.1 The test aerosol shall be poly disperse solid-phase KCL particle generated
from an aqueous solution. The aerosol generator shall provide a stable challenge
aerosol of sufficient concentration over the 0.3 to 10µ diameter size range to
meet the requirements of section 10 without overloading the aerosol particle
4.3.2 The aerosol generation system shall be designed to ensure that the KCL
particles are dry prior to being introduced into the test duct. After drying,
the aerosol must be electrostatic charge-neutralized by passing it through an
aerosol neutralizer with a Kr85 radioactive source.
4.4.1 The upstream and downstream sample lines must be made of rigid electrical
ly-grounded metallic tubing. The use a short length (2" max.) of straight,
flexible, "electrical ly-dissapative" tubing to make the final
connection to the particle counter is acceptable.
4.6 Particle Counter
4.6.1 The particle counter shall be capable of counting and sizing individual
KCL particles in the 0.3 to 10µ diameter size range "The counting
efficiency shall be 50% min. for 0.3µ KCL particles".
4.6.2 The particle counter shall measure and then group the test aerosol
particles in 12 size ranges. The range boundaries, based on the physical size of
the KCL aerosol, shall conform to the table 4-1. The particle counter's
correlation of measured response voltage to physical size shall be monotonic for
KCL particles from 0.3 to 10µ such that only one size range shall be indicated
for any measured response.
4.6.3 The primary size calibration of the particle counter shall be to report
the actual physical size of the KCL particle. A secondary calibration shall be
performed on the particle counter using PSL particles. The secondary calibration
data shall be converted to primary calibration data. "The particle counter
manufacturer shall determine the secondary to primary conversion for the
instrument type by calibrating an instrument of that type with both mono
disperse KCL and PSL. The change in indicated particle size when making the
secondary to primary conversion shall not exceed 20%." Exception: "An
exception to 4.6.3 is permitted for light-scattering particle counters. If this
exception is taken, the manufacturer shall determine the secondary to primary
conversion by calculating the MIE scattering response for KCL and PSL, assuming
spherical particles and considering the light source and collection angles of
the instrument. The change in indicated particle size when making the secondary
to primary conversion shall not exceed 20%".
4.6.4 The secondary calibration shall be performed at least once a year. The
primary calibration shall then be updated using the conversion calculation and
the secondary calibration.
4.6.5 A minimum of 90% of all observed counts shall register in size ranges 1
and 2 of table 4-1 when the particle counter is challenged with mono dispersed
0.40µ diameter particles. Secondary calibration data shall be used to
demonstrate compliance with this requirement. Refer to 4.6.4.
Table 4-1 Particle Counter Size Range Boundaries
Size Range Lower
Size Range Upper
Particle Size (microns)
4.6.6 The particle counter shall have less than
10% coincidence loss at a particle counting rate of 300,000 particles per minute
and shall have a minimum viewed volume flow rate of 0.10 cfm. This flow rate
shall not change with a 4.0" of water in the pressure of sampled air.
4.6.7 The measured particle concentration shall be a maximum 5.66 particles/ft3
when the particle counter is sampling air with a HEPA or ULPA filter on it's
5 Apparatus Qualification Testing
5.1 Apparatus qualification tests shall verify quantitatively that the test rig
and sampling procedures are capable of providing reliable particle size
efficiency measurements. The tests shall be performed in accordance with table
5.3 Aerosol concentration uniformity in the test duct
5.3.1 The uniformity of the challenge aerosol concentration across the duct
crosssection shall be determined by a nine point traverse by using the grid
points as shown in figure 5-1. The traverse shall be made by either (A)
installing nine sample probes of identical curvature, diameter, and inlet nozzle
diameter but of variable lengths or (B) repositioning a single probe. The inlet
nozzle of the sample probe(s) shall be sharp edged and shall maintain isokinetic
sampling within 10 % at 1990 cfm. The same nozzle diameter shall be used at all
5.3.2 The aerosol concentration measurements shall be made with a particle
counter meeting the specifications of 4.6. A one minute sample shall be taken at
each grid point with the aerosol generator operating. After sampling all nine
points the traverse shall be repeated four more times to provide a total of five
samples from each point. The five samples shall then be averaged for each of the
12 particle size ranges. The traverse measurements shall be performed at air
flows of 510, 1990, and 3180 cfm.
5.3.3 The CV of the corresponding nine grid point particle concentrations shall
be less than 10% for each air flow in each of the 12 particle counter size
5.4 Downstream Mixing of Aerosol
5.4.1 A mixing test shall be performed to ensure that all aerosol that
penetrates the air cleaner is detectable by the downstream sampler. The mixing
test shall be performed at air flows of 510, 1990, and 3180 cfm. The point of
aerosol injection immediately downstream of the device section shall be
traversed and the downstream sampling probe shall remain stationary in it's
normal center-of -duct sampling location.
5.4.2 A HEPA filter shall be installed to obtain a smooth airflow at the output
of the device section. An aerosol nebulizer shall nebulize a KCI/water solution
( prepared using a ratio of 300g of KCL to 1 1000ml water ) into an aerosol of
primarily submicron sizes. A rigid extension tube with a length sufficient to
reach each of the injection points shall be affixed to the nebulizer outlet. A
900 bend shall be placed at the outlet ofthe tube to allow injection of the
aerosol in the direction of the airflow. The injection probe shall point
downstream. The aerosol shall be injected immediately downstream (within
10" ) of the HEPA filter at pre selected points located around the
perimeter of the test duct and at the center of the duct as indicated in figure
5-2. The flow rate through the nebulizer and the diameter of the injection tube
outlet shall be adjusted to provide an injection air velocity within + 50% of
the mean duct velocity.
5.43 Sampling sequence: A one minute sample from the downstream probe shall be
acquired with the nebulizer operating and the injection tube positioned at the
'first injection grid point. The injection point shall then be moved 'to the
next grid point location. A new one minute sample shall be obtained after
waiting a minimum of 30 seconds. The procedure shall be repeated until all nine
grid points have been sampled.
5.4.4 The aerosol injection traverse shall be repeated two more times to provide
triplicate measurements at each grid point.
5.4.5 The downstream aerosol concentration shall be measured as total aerosol
concentration > 0.3um. The CV of the corresponding nine downstream grid point
particle concentrations must be less than 5% for each airflow.
5.5 Aerosol Generator Response Time
5.5.1 Measure 'the time interval for 'the aerosol concentration to go from
background level to steady test level. The test shall be performed at an airflow
rate of 1990 cfm with 'the particle counter sampling from the upstream probe.
Similarly, measure the time interval for the aerosol to return to background
level after turning of the generator.
5.5.2 Record 'the time interval needed for the aerosol concentration to reach
it's steady test value after being activated, and also record the time interval
concentration to return to background level after deactivating the aerosol
generating system. These time intervals shall be used as the minimum waiting
time between (a) activating the aerosol generator and beginning the particle
counter sampling sequence and (b) deactivating the aerosol generator and
beginning the particle counter sampling sequence for determination of background
aerosol concentrations, respectively.
5.6 Concentration Limit of the Particle Counter
5.6.1 A series of controlled 50% dilution tests shall be performed over a range
of challenge aerosol concentrations to ensure that the concentration used in the
tests does not overload the particle counter. The lowest concentration level
shall be less than 1% of the instrument's stated concentration limit. The tests
shall be performed by one of the following methods: (a) Achieve the 50% dilution
by alternating the test duct airflow from 996 to 1990 CFM while operating the
aerosol generator at a constant output. The aerosol samples shall be obtained
from the upstream sampling location only. (b) A controlled 50% dilution of the
sample pie stream to the particle counter.
5.6.3 The tests shall be performed over a sufficient range of challenge
concentration levels to demonstrate 'that the particle counter is not overloaded
at the intended test concentration.
5.6.4 A minimum of three samples shall be obtained for the undiluted and 50%
dilution levels. These samples shall be obtained by alternately sampling the
undiluted and 50% diluted level. The dilution ratio shall be computed as the
ratio of the average diluted concentration to the average undiluted
concentration. The dilution ratio calculation shall be based only on the 0.30 to
0.40um channel of the particle counter. The measured dilution ratio in the
absence of coincidence error shall be within 10 percentage points of the
expected 50% dilution.
5.7 100% Efficiency Test
5.7.1 An efficiency test shall be performed using a HEPA or ULPA filter as a
test device to ensure that the test duct and sampling system are capable of
providing a 100% efficiency measurement. The test procedures for determination
of PSE given in Section 10 shall be followed, and the "test shall be
performed at an airflow of 1990 cfm .
5.7.2 The computed PSE values shall be greater than 99% for all particle sizes.
5.8 0% Efficiency Test
5.8.1 An efficiency test shall be performed
without a test device in place to check the adequacy of the duct, sampling,
measurement and aerosol generation systems.
5.8.2 The test procedures for determination of PSE given in Section 10 shall be
5.8.3 A total of fifteen 0% efficiency tests shall be performed; five at each of
three flow rates: 510, 1990, and 3180 cfm. The average of the five efficiency
values at each particle size shall be computed for each flow rate. The average
efficiency for each particle size shall be between -10% and 20% for all particle
5.10 Particle Counter Zero
5. 10. 1 The zero count of the particle counter shall be verified to be < 10
total counts per minute in the 0.30 to 10um size range when operating with a
high efficiency filter attached directly to the instrument's inlet.
5.11 Particle Counter Sizing Accuracy
5.11.1 The sizing accuracy of the particle counter shall be checked by sampling
an aerosol containing monodisperse polystyrene spheres of a known size. A
relative maximum particle count shall appear in the particle counter sizing
channel which encompasses the PSL diameter.
The remainder of the standard consists of tests for airflow, filter
efficiency, dust loading, etc.