Accessed from 219.93.4.

130 by global5 on Tue Mar 04 22:19:27 EST 2014

5718 〈467〉 Residual Solvents / Chemical Tests First Supplement to USP 36–NF 31

A factor of 10 to account for variability between individu- A3.1. Values Used in the Calculations in This Document
als. A factor of 10 is generally given for all organic sol-
Rat body weight 425 g
F2 = vents, and 10 is used consistently in this general chapter.
Pregnant rat body weight 330 g
Mouse body weight 28 g
A variable factor to account for toxicity studies of short- Pregnant mouse body weight 30 g
F3 = term exposure Guinea-pig body weight 500 g
1 for studies that last at least one half-lifetime Rhesus monkey body weight 2.5 kg
(1 year for rodents or rabbits; 7 years for Rabbit body weight (pregnant or not) 4 kg
F3 = cats, dogs, and monkeys) Beagle dog body weight 11.5 kg
1 for reproductive studies in which the whole Rat respiratory volume 290 L/day
F3 = period of organogenesis is covered
Mouse respiratory volume 43 L/day
2 for a 6-month study in rodents, or a 3.5-
Rabbit respiratory volume 1440 L/day
F3 = year study in nonrodents
Guinea-pig respiratory volume 430 L/day
5 for a 3-month study in rodents, or a 2-year
F3 = study in nonrodents Human respiratory volume 28,800 L/day
F3 = 10 for studies of a shorter duration Dog respiratory volume 9000 L/day
Monkey respiratory volume 1150 L/day
In all cases, the higher factor has been used for study dura- Mouse water consumption 5 mL/day
tions between the time points (e.g., a factor of 2 for a Rat water consumption 30 mL/day
9-month rodent study). Rat food consumption 30 g/day

A factor that may be applied in cases of severe toxicity, e. The equation for an ideal gas, PV = nRT, is used to con-
g., nongenotoxic carcinogenicity, neurotoxicity, or ter- vert concentrations of gases used in inhalation studies from
atogenicity. In studies of reproductive toxicity, the fol- units of ppm to units of mg/L or mg/m3. Consider as an
F4 = lowing factors are used: example the rat reproductive toxicity study by inhalation of
1 for fetal toxicity associated with maternal carbon tetrachloride (molecular weight 153.84) summarized
F4 = toxicity in Pharmeuropa, Vol. 9, No. 1, Supplement, April 1997,
F4 = 5 for fetal toxicity without maternal toxicity page S9.
5 for a teratogenic effect with maternal
F4 = toxicity
10 for a teratogenic effect without maternal
F4 = toxicity
The relationship 1000 L = 1 m3 is used to convert to mg/
m3.
A variable factor that may be applied if the no-effect level
F5 = was not established

When only a LOEL is available, a factor of up to 10 can be

used, depending on the severity of the toxicity. The weight
adjustment assumes an arbitrary adult human body weight
for either sex of 50 kilograms (kg). This relatively low weight
Physical Tests and
provides an additional safety factor against the standard
weights of 60 kg or 70 kg that are often used in this type of
Determinations
calculation. It is recognized that some adult patients weigh
less than 50 kg; these patients are considered to be accom-
modated by the built-in safety factors used to determine a
PDE. If the solvent was present in a formulation specifically
intended for pediatric use, an adjustment for a lower body
weight would be appropriate.
As an example of the application of this equation, con-
sider a toxicity study of acetonitrile in mice that is summa-
〈643〉 TOTAL ORGANIC CARBON
rized in Pharmeuropa, Vol. 9, No. 1, Supplement, April
1997, page S24. The NOEL is calculated to be 50.7 mg kg–1
day–1. The PDE for acetonitrile in this study is calculated as
follows: Change to read:
PDE = (50.7 mg kg −1 day × 50 kg)/(12 × 10 × 5 × 1 × 1) =
−1
Total organic carbon (TOC) is an indirect measure of or-
4.22 mg day−1 ganic molecules present in pharmaceutical waters measured
as carbon. Organic molecules are introduced into the water
In this example, from the source water, from purification and distribution
system materials, ■from biofilm growing in the system, and
F1 = 12 to account for the extrapolation from mice to humans from the packaging of sterile and nonsterile waters.■1S (USP36)
F2 = 10 to account for differences between individual humans TOC can also be used as a process control attribute to mon-
F3 = 5 because the duration of the study was only 13 weeks itor the performance of unit operations comprising the puri-
F4 = 1 because no severe toxicity was encountered fication and distribution system. A TOC measurement is not
a replacement test for endotoxin or microbiological control.
F5 = 1 because the no-effect level was determined
Although there can be a qualitative relationship between a
food source (TOC) and microbiological activity, there is no
direct numerical correlation.
A number of acceptable methods exist for analyzing TOC.
This chapter does not endorse, limit, or prevent any tech-

Official from December 1, 2013

Copyright (c) 2014 The United States Pharmacopeial Convention. All rights reserved.
Accessed from 219.93.4.130 by global5 on Tue Mar 04 22:19:27 EST 2014
First Supplement to USP 36–NF 31
nologies from being used, but this chapter provides guid-
ance on how to qualify these analytical technologies for use
as well as guidance on how to interpret instrument results
for use as a limit test.
Apparatuses commonly used to determine TOC in water
for pharmaceutical use have in common the objective of
oxidizing the organic molecules in the water to produce car-
bon dioxide followed by the measurement of the amount of
carbon dioxide produced. Then the amount of CO2 pro-
duced is determined and used to calculate the organic car-
bon concentration in the water.
All technologies must discriminate between the inorganic
carbon, which may be present in the water from sources
such as dissolved CO2 and bicarbonate, and the CO2 gener-
ated from the oxidation of organic molecules in the sample.
The discrimination may be accomplished either by deter-
mining the inorganic carbon and subtracting it from the
total carbon (total carbon is the sum of organic carbon and
inorganic carbon), or by purging inorganic carbon from the
sample before oxidation. Although purging may entrain or-
ganic molecules, such purgeable organic carbon is present
in negligible quantities in water for pharmaceutical use.
Change to read:
■ BULK WATER
The following sections apply to tests for bulk Purified
Water, Water for Injection, Water for Hemodialysis, and the
condensate of Pure Steam.■1S (USP36)
Apparatus Requirements: This test method is performed
either as an on-line test or as an off-line laboratory test us-
ing a calibrated instrument. The suitability of the apparatus
must be periodically demonstrated as described below. In
addition, it must have a manufacturer’s specified limit of
detection of 0.05 mg/L (0.05 ppm) or lower of carbon.
When testing water for quality control purposes, ensure
that the instrument and its data are under appropriate con-
trol and that the sampling approaches and locations of both
on-line and off-line measurements are representative of the
quality of the water used. The nature of the water produc-
tion, distribution, and use should be considered when se-
lecting either on-line or off-line measurement.
USP Reference Standards 〈11〉: USP 1,4-Benzoquinone RS.
USP Sucrose RS.
Reagent Water: Use water having a TOC level of not
more than 0.10 mg/L. [NOTE—A conductivity requirement
may be necessary in order to ensure method reliability.]
Container Preparation: Organic contamination of con-
tainers results in higher TOC values. Therefore, use labware
and containers that have been scrupulously cleaned of or-
ganic residues. Any method that is effective in removing
organic matter can be used (see Cleaning Glass Apparatus
〈1051〉). Use Reagent Water for the final rinse.
Standard Solution: Unless otherwise directed in the indi-
vidual monograph, dissolve in the Reagent Water an accu-
rately weighed quantity of USP Sucrose RS to obtain a solu-
tion having a concentration of 1.19 mg/L of sucrose
(0.50 mg/L of carbon).
System Suitability Solution: Dissolve in Reagent Water an
accurately weighed quantity of USP 1,4-Benzoquinone RS to
obtain a solution having a concentration of 0.75 mg/L
(0.50 mg/L of carbon).
Reagent Water Control: Use a suitable quantity of Rea-
gent Water obtained at the same time as that used in the
preparation of the Standard Solution and the System Suitabil-
ity Solution.
Water Sample: Obtain an on-line or off-line sample that
suitably reflects the quality of water used.
Official from December 1, 2013
Copyright (c) 2014 The United States Pharmacopeial Convention. All rights reserved.
Physical Tests / 〈643〉 Total Organic Carbon 5719
Other Control Solutions: Prepare appropriate reagent
blank solutions or other specified solutions needed for estab-
lishing the apparatus baseline or for calibration adjustments
following the manufacturer’s instructions, and run the ap-
propriate blanks to zero the instrument, if necessary.
System Suitability: Test the Reagent Water Control in the
apparatus, and record the response, rW. Repeat the test us-
ing the Standard Solution, and record the response, rS. Cal-
culate the corrected Standard Solution response, which is
also the limit response, by subtracting the Reagent Water
Control response from the response of the Standard Solution.
The theoretical limit of 0.50 mg/L of carbon is equal to the
corrected Standard Solution response, rS − rW. Test the System
Suitability Solution in the apparatus, and record the re-
sponse, rSS. Calculate the corrected System Suitability Solution
response by subtracting the Reagent Water Control response
from the response of the System Suitability Solution, rSS − rW.
Calculate the percent response efficiency for the System Suit-
ability Solution:
% response efficiency = 100[(rSS − rW)/(rS − rW)]
where rSS is the instrument response to the System Suitability
Solution; rW is the instrument response to the Reagent Water
Control; and rS is the instrument response to the Standard
Solution. The system is suitable if the percent response effi-
ciency is not less than 85% and not more than 115%.
Procedure: Perform the test on the Water Sample, and re-
cord the response, rU. The Water Sample meets the require-
ments if rU is not more than the limit response, rS − rW. This
method can be performed using on-line or off-line instru-
mentation that meets the Apparatus Requirements.
Add the following:
■ STERILE WATER
The following sections apply to tests for Sterile Water for
Injection, Sterile Purified Water, Sterile Water for Irrigation, and
Sterile Water for Inhalation.
Follow the requirements in Bulk Water, with the following
exceptions.
Apparatus Requirements: In addition to the Apparatus
Requirements in Bulk Water, the apparatus must have a man-
ufacturer’s specified limit of detection of 0.10 mg/L
(0.10 ppm) or lower of carbon.
Reagent Water: Use water having a TOC level of not
more than 0.50 mg/L. [NOTE—A conductivity requirement
may be necessary in order to ensure method reliability.]
Standard Solution: Unless otherwise directed in the in-
dividual monograph, dissolve in the Reagent Water an accu-
rately weighed quantity of USP Sucrose RS to obtain a solu-
tion having a concentration of 19.0 mg/L of sucrose
(8.0 mg/L of carbon).
System Suitability Solution: Dissolve in Reagent Water
an accurately weighed quantity of USP 1,4-Benzoquinone RS
to obtain a solution having a concentration of 12.0 mg/L
(8.0 mg/L of carbon).
Water Sample: Obtain a sample that suitably reflects
the quality of water used. Before opening, vigorously agitate
the package to homogenize the water sample. Several pack-
ages may be required in order to collect sufficient water for
analysis.
System Suitability: Test the Reagent Water Control in
the apparatus, and record the response, rW. Repeat the test
using the Standard Solution, and record the response, rS.
Calculate the corrected Standard Solution response, which is
also the limit response, by subtracting the Reagent Water
Control response from the response of the Standard Solution.
The theoretical limit of 8.0 mg/L of carbon is equal to the
corrected Standard Solution response, rS − rW. Test the System
 Accessed from 219.93.4.130 by global5 on Tue Mar 04 22:19:27 EST 2014
5720 〈643〉 Total Organic Carbon / Physical Tests
Suitability Solution in the apparatus, and record the re-
sponse, rSS. Calculate the corrected System Suitability Solution
response by subtracting the Reagent Water Control response
from the response of the System Suitability Solution, rSS − rW.
Calculate the percent response efficiency for the System Suit-
ability Solution:
% response efficiency = 100[(rSS − rW)/(rS − rW)]
where rSS is the instrument response to the System Suitability
Solution; rW is the instrument response to the Reagent Water
Control; and rS is the instrument response to the Standard
Solution. The system is suitable if the percent response effi-
ciency is not less than 85% and not more than 115%.
Procedure: Perform the test on the Water Sample, and
record the response, rU. The Water Sample meets the re-
quirements if rU is not more than the limit response, rS − rW,
determined in the System Suitability requirements in Sterile
Water.■1S (USP36)
〈645〉 WATER CONDUCTIVITY
Electrical conductivity in water is a measure of the ion-
facilitated electron flow through it. Water molecules dissoci-
ate into ions as a function of pH and temperature and result
in a very predictable conductivity. Some gases, most nota-
bly carbon dioxide, readily dissolve in water and interact to
form ions, which predictably affect conductivity as well as
pH. For the purpose of this discussion, these ions and their
resulting conductivity can be considered intrinsic to the
water.
Water conductivity is also affected by the presence of ex-
traneous ions. The extraneous ions used in modeling the
conductivity specifications described below are the chloride
and sodium ions. The conductivity of the ubiquitous chlo-
ride ion (at the theoretical endpoint concentration of
0.47 ppm when it was a required attribute test in USP XXII
and earlier revisions) and the ammonium ion (at the limit of
0.3 ppm) represent a major portion of the allowed water
impurity level. A balancing quantity of cations, such as so-
dium ions, is included in this allowed impurity level to main-
tain electroneutrality. Extraneous ions such as these may
have significant impact on the water’s chemical purity and
suitability for use in pharmaceutical applications. The proce-
dure in the section Bulk Water is specified for measuring the
conductivity of waters such as Purified Water, Water for Injec-
tion, Water for Hemodialysis, and the condensate of Pure
Steam. The procedure in the section Sterile Water is specified
for measuring the conductivity of waters such as Sterile Puri-
fied Water, Sterile Water for Injection, Sterile Water for Inhala-
tion, and Sterile Water for Irrigation.
Online conductivity testing provides real-time measure-
ments and opportunities for real-time process control, deci-
sion, and intervention. Precaution should be taken while col-
lecting water samples for off-line conductivity
measurements. The sample may be affected by the sam-
pling method, the sampling container, and environmental
factors such as ambient carbon dioxide concentration and
organic vapors.
INSTRUMENT SPECIFICATIONS AND
OPERATING PARAMETERS
Water conductivity must be measured accurately using
calibrated instrumentation. The conductivity cell constant, a
factor that represents the geometrical properties of the con-
Official from December 1, 2013
Copyright (c) 2014 The United States Pharmacopeial Convention. All rights reserved.
First Supplement to USP 36–NF 31
ductivity sensor, must be known within ±2%. The cell con-
stant can be verified directly by using a solution of known
or traceable conductivity, or indirectly by comparing the in-
strument reading taken with the conductivity sensor in
question to readings from a conductivity sensor of known or
traceable cell constant.
Meter calibration is accomplished by replacing the con-
ductivity sensor with NIST (or equivalent local national au-
thority)-traceable precision resistors (accurate to ±0.1% of
the stated value) or an equivalently accurate adjustable resis-
tance device, such as a Wheatstone Bridge, to give a pre-
dicted instrument response. Each scale on the meter may
require separate calibration prior to use. The frequency of
recalibration is a function of instrument design, degree of
use, etc. However, because some multiple-scale instruments
have a single calibration adjustment, recalibration may be
required between each use of a different scale. Excluding
the conductivity sensor cell constant accuracy, the instru-
ment accuracy must be ±0.1 µS/cm.
In order to increase the measurement accuracy on the
conductivity ranges used, which can be large, and to ensure
a complete equipment calibration, it is suggested that peri-
odic verification of the entire equipment be performed. This
could be done by comparing the conductivity/resistivity val-
ues displayed by the measuring equipment with those of an
external calibrated conductivity-measuring device. The two
nontemperature-compensated conductivity or resistivity val-
ues must be equivalent to within ±20% of each other, or at
a difference that is acceptable on the basis of product water
criticality and/or the water conductivity ranges in which the
measurements are taken. The two conductivity sensors
should be positioned close enough together to measure the
same water sample in the same environmental conditions.
In addition to the verification method performed in
nontemperature-compensated mode, a similar verification
performed in temperature-compensated mode could be per-
formed to ensure an appropriate accuracy of the equipment
when such a mode is used for trending or other purposes.
Because temperature has a substantial impact on conduc-
tivity readings of specimens at high and low temperatures,
many instruments automatically correct the actual reading
to display the value that theoretically would be observed at
the nominal temperature of 25°. This is typically done using
a temperature sensor embedded in the conductivity sensor
and an algorithm in the instrument’s circuitry. This tempera-
ture compensation algorithm may not be accurate. Conduc-
tivity values used in this method are nontemperature-com-
pensated measurements. Temperature measurement is
required for the performance of the Stage 1 test. It may be
made using the temperature sensor embedded in the con-
ductivity cell sensor. An external temperature sensor posi-
tioned near the conductivity sensor is also acceptable. Accu-
racy of the temperature measurement must be ±2°.
The procedures below shall be performed using instru-
mentation that has been calibrated, has conductivity sensor
cell constants that have been accurately determined, and
has a temperature compensation function that has been dis-
abled. For both online and offline measurements, the suita-
bility of instrumentation for quality control testing is also
dependent on the sampling location(s) in the water system.
The selected sampling instrument location(s) must reflect
the quality of the water used.
BULK WATER
The procedure and test limits in this section are intended
for Purified Water, Water for Injection, Water for Hemodialysis,
the condensate of Pure Steam, and any other monographs
which specify this section.
The combined conductivities of the intrinsic and extrane-
ous ions vary as a function of pH and are the basis for the
conductivity specifications described in the Stage 3—pH and
Conductivity Requirements table and used when performing
Stage 3 of the test method. Two preliminary stages are in-