American Journal of Analytical Chemistry, 2013, 4, 776-780
Published Online December 2013 (http://www.scirp.org/journal/ajac)
http://dx.doi.org/10.4236/ajac.2013.412095
Open Access AJAC
Ammonia Removal from Rodent Habitat Operations in
Space Using Phosphoric Acid Treated Activated Carbon
Zhe Lu1*, Jacob A. Hines2*, Daniel J. Rozewicz1, Michael L. Hines3
1Lockheed Martin Space OPNS, Moffett Field, USA
2University of California, Santa Barbara, USA
3NASA Ames Research Center, Moffett Field, USA
Received October 30, 2013; revised November 29, 2013; accepted December 15, 2013
Copyright © 2013 Zhe Lu et al. This is an open access article distributed under the Creative Commons Attribution License, which
permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ABSTRACT
To accommodate long duration biology research with rodent habitats on the International Space Station while providing
a healthy living and working environment for crewmembers, NASA Ames Research Center developed a new exhaust
filter for odor control for the Animal Enclosure Module (AEM), which houses mice and rats. The new exhaust filter
uses activated carbon pellets as adsorbents, with phosphoric acid (H3PO4) impregnated on the surface. The deodoriza-
tion performance of the new exhaust filters for AEM units housed with mice was evaluated. The ammonia breakthrough
time of the exhaust filters was also investigated. The results indicated that H3PO4 treated activated carbon exhibited a
high ammonia adsorption capacity of more than 90%. Furthermore, the new exhaust filter can effectively control the
odor from the AEM units for a 45-day (minimum) flight mission with a given animal biomass.
Keywords: Activated Carbon; Impregnation; Adsorption; Ammonia; International Space Station
1. Introduction
For centuries, humans have observed animals in order to
better understand aspects of human biology. In modern
times, animal research has become the gold standard for
basic biology and medicine. With animal subjects, it is
possible to control and reproduce environmental condi-
tions, experimental subjects, and protocol, a set of ad-
vantages that is often difficult if not impossible to achieve
with human subjects. The value of animal research ap-
plies equally well to research in space as it does on the
ground. Space animal research is essential for under-
standing the impacts of spaceflight on physiological sys-
tems and for the development of therapies that will miti-
gate detrimental responses to spaceflight. The rodent, for
example, is an ideal surrogate for establishing the tem-
poral baseline effects of long term exposure to space-
flight.
The National Aeronautics and Space Administration
(NASA) Animal Enclosure Module (AEM), designed to
provide a flight habitat environment for rodents, has been
routinely used for spaceflight to conduct rodent micro-
gravity research studies in space. The rodents in the
AEM are a source of numerous airborne contaminants.
Odor complaints that have been risen about the rodents
involve both the unpleasant character of the odor, as well
as the potential adverse health effects.
Although little attempt has been made to identify the
chemical composition of the exhaust airstream from the
AEM, over 100 chemical compounds have been identi-
fied in the airstream from various animal houses [1-4].
Ammonia is one of the major odor causing compounds in
animal houses and presents the greatest risk to the envi-
ronment. It is produced by the decomposition of nitro-
gen-containing compounds in the excreta, especially in
the urine. Levels of 50 - 100 ppm of ammonia can cause
irritation of eyes, throat, and nose, and has a detection
threshold between 5 and 18 ppm for humans [5]. Ammo-
ia levels for a well-ventilated animal house are in the
range of 5 to 10 ppm as set by the US Occupational
Safety and Health Administration (OSHA) [6]. The Space-
craft Maximum Allowance Concentration (SMAC) of am-
monia in the International Space Station (ISS), set by the
NASA/JSC Toxicology Group, is 30 ppm for less than 1
hour of exposure and 10 ppm for an exposure of more
than 7 days [7].
*Corresponding authors. NASA Ames Research Center (ARC) has undertaken

Z. LU ET AL. 777
the effort to build a new lot of exhaust filters for odor
control, especially for ammonia removal, for upcoming
long-term science payload missions with AEM payloads.
Many approaches have been tested to mitigate ammo-
nia, but most of them remain on laboratory scales. These
techniques include absorption by solutions, separation
using membranes, catalytic decomposition, and adsorp-
tion by porous solids [8,9]. However, the adsorption on
porous solids will be an excellent option to remove am-
monia in the flight habitat environment, as liquid solu-
tions are difficult to maintain in space. The ammonia
adsorbents are mainly zeolite, activated carbon, and acti-
vated alumina. The ammonia adsorption on porous solids
can take place through physisorption and chemisorption
processes [10]. The chemisorption process is usually
stronger than physisorption, because the chemisorption
process involves a chemical reaction between ammonia
and the adsorbents, while the physisorption process re-
lates to a pore’s filling being driven by weak van de
Waals forces.
Among the potential ammonia adsorbents, activated
carbon is commonly used. Activated carbon exhibits high
surface area and highly developed porous structures,
which facilitate physical adsorption. Most importantly,
strong interactions between the gas and the adsorbent can
be enabled by tailoring the porous structure and surface
chemistry of activated carbon [8,11,12]. The alterations
of the activated carbon can be accomplished by impreg-
nation of inorganic compounds, which are anticipated to
develop stronger interactions between ammonia and the
surface of the activated carbon [13-15]. These impreg-
nants react instantaneously with ammonia in air that has
been filtered by previous sorbent layers to remove mois-
ture to form stable chemical compounds that are ire-
versibly bound to the media as inorganic [16]. The prop-
erties of activated carbon will be optimized by the im-
pregnation of suitable chemicals on the internal surface
for the chemisorption of certain gases.
The objective of this paper is to present the compre-
hensive design and configuration of the long-term odif-
erous organic compound filter, which uses phosphoric
acid loaded activated carbon as its adsorbent to remove
odorants, especially ammonia. Efforts were made to meas-
ure the effectiveness of the long duration exhaust filter
designed for odor control. The deodorization perform-
ance of the exhaust filters is examined based on 45-day
(or longer) science missions in space with mice-loaded
AEM units. The lifetime of the exhaust filter will also be
explored by testing the time until the odor breakthrough
of the exhaust filter.
2. Experimental
2.1. Materials
New exhaust filters (Figure 1) for the AEM have been
assembled and built at NASA Ames Research Center.
The exhaust filter is 21.9 cm× 35.5 cm× 6.5 cm and the
flowrate of theairstream from the AEM unitto the ex-
haust filter was controlled at 0.1 L/sec.
The exhaust filter in the AEM is designed to prevent
the escape of particulate matter into the cabin atmosphere,
as well as contain animal odor and neutralize urine within
the AEM. The exhaust filter, rated for 45 days of odor
control, is made up of 9 layers including Bondina, fabric
sorbents, activated carbon, filtrate, and zeolite, as shown
in Figure 2. Layer 1 directly exposed to the rodent cham-
ber. This layer, as well as layer 4, consists of phosphoric
acid (H3PO4) treated Bondina with 20 g pre-loaded H3PO4
impregnated on the surface of Bondina for removal of
alkaline gas, such as NH3 and amines. Layers 2, 3, 5, and
7 consist of nonwoven fabric with evenly distributed
holes for airflow, provide excellent liquidretention and
Figure 1. Picture of the exhaust filter for the Animal En-
closure Module (AEM).
Airflow
AirGap AirGap
123456789
1) Bondina with acid; 2) Sorbent, yellow, punched, with acid; 3) Sorbent,
yellow, punched; 4) Bondina with acid; 5) Sorbent, yellow, punched; 6)
Zeolite; 7) Universal Sorbent, yellow, punched; 8) Activated Carbon Bed,
with acid; 9) G200 Filtrete.
Figure 2. Diagram of the exhaust filter layers for the Ani-
mal Enclosure Module (AEM).
Open Access AJAC

Z. LU ET AL.
778
high air flow rate, and are used to stop urine. Air gaps
between layers 2 and 3, and layers 7 and 8, were de-
signed to separate fabric filter layers and allow a fast
airflow through the layers. Layer 7 contains zeolite pel-
lets from Lewcott Corp, Millbury, MA, which attempt to
facilitate the adsorption of both urine and ammonia.
Layer 8 is the carbon bed, which is packed with com-
mercially available Ammosorb (NUCON International,
Columbus, OH), a pelleted, activated carbon with phos-
phoric acid loaded on the surface to remove alkali gases
such as ammonia and amine vapors. The last layer of the
exhaust filter is Trion G200 filtrete, which will help to
reduce dust, mold spore and odor in the air stream.
Three AEM units (AEM 101, 102, and 103) were
loaded with an exhaust filter, mouse food bars, and dis-
tilled water, and were used for mice habitat. One AEM
unit (AEM 104), had this same setup and was used as a
control chamber for the odor evaluation. Ten (10) mice,
with 5 mice on each side of the split chamber, were
loaded in the three AEM units (101, 102, and 103). All of
the four AEM units were instrumented with temperature
probes and a gas port for periodic sampling of ammonia
(NH3), and daily hardware and animal health checks were
performed.
The AEM exchanges chamber air with the exhaust fil-
ter, a process wherein the air from the rodent cage, pre-
viously drawn in through the inlet filter, flows through
the exhaust filter prior to exiting the AEM. High effi-
ciency air filters prevent the escape of particulate matter
into the cabin atmosphere, and treated activated carbon-
inside the filters helps contain animal odor and neutralize
urine within the AEM. After exiting the habitat through
the exhaust filter, the filtered air is drawn through the
exhaust fans into the cabin.
Gas samples were collected from the gas port that di-
rectly connected to the AEM main chamber. Concentra-
tions of NH3 were measured from test day 21 onwards
using two Dräger test pumps with a detection range of
0.20 - 5.0 ppm and 5 - 70 ppm, respectively.
2.2. Odor Measuring Method
The odor expelled from the four investigated AEM units
was evaluated using panelists to rank samples. A group
of volunteers were trained to conduct the daily odor
evaluations. An arbitrary scale was used to describe the
intensity of the odor from the AEM units. Each AEM
unit was covered with a blue shroud to prevent ambient
light from entering cage environment and prevented odor
panelists from viewing the cage interior.
During the evaluations, sniffers were asked to stand a
distance of 6 - 8 in from the AEM exhaust outlet and
scored the odor using a 5-point scoring system: 0 = un-
detectable, 1 = barely detectable, 2 = easily detectable, 3
= objectionable (disagreeable), and 4 = revolting (ex-
tremely offensive). Group training, instruction, and cali-
bration were provided to evaluators prior to the tests.
Score selections were based on odor interpretation with
respect to predetermined and formerly presented (sniffed)
standards for each level of the scoring system. Average
odor scores for each test day were then calculated.
The animal tests will be stopped when the average
odor scores are over 3.0 for two consecutive days, the
mouse food bars need to be replaced, or when the mice
are observed to be in bad health, whichever appears first.
The breakthrough time of the new exhaust filter will be
identified for AEM units 102, 103, and 104 when the
average odor scores go over 3.0 for two consecutive
days.
3. Results and Discussions
Figure 3 shows the odor evaluation results from the
AEM units on test days T + 0 through T + 50. The total
number of people in the panel was no less than 12 for
any test day. At the beginning of the test (T + 0), there
was an odor reading of 1.4 for the control AEM unit 104,
and 1.05, 1.4, and 1.6 for AEM unit 101, 102, and 103,
respectively. This indicated that some individuals identi-
fied a background odor, which waslikelythe odor of the
mouse food bars, and that the exhaust filters with H3PO4
treated activated carbon did not effectively remove the
odorants in the mouse food bars. We could not comment
on the reactions of odorants from mouse food bars against
the sorbent materials in the exhaust filter in the present
work, as the odor causing compounds in the mouse food
bars have not been identified yet.
From test day 7 to 50, the odor scores of the control
AEM unit decreased relative to the first test day and
varied in the range of 0.29 - 0.88, likely as a result of the
dehydration of the mouse food bars with operation time.
As indicated Figure 3, the average odor scores of animal-
loaded AEM units slightly decreased on day 7 for AEM
0
1
2
3
4
161116 21 26 31 36 41 46
Averag e Pa n eli st Od or S co re (0 -4 )
Test Day
Average Panelist Odor Score vs Time
AEM 101 Odor ScoreAEM 102 Odor Score
AEM 103 Odor ScoreAEM 104
(
c
)
Odor Score
Odor Breakthrough
Figure 3. Odor evaluation results of the AEM units.
Open Access AJAC

Z. LU ET AL.
Open Access AJAC
779
units 102, 103, and 104, and then increased gradually
from day 14 to day 50. This can be attributed to the fact
that the odorants from the airstream of the AEM units
were removed by the exhaust filter, and that the odor
containment performance of the exhaust filter decreased
with longer operation duration. The decreased odor con-
trol performance of the exhaust filter could be either due
to the decrease of active sites, or the increased penetra-
tion of other odorants that were not absorbed by the filter
with extended operation time.
Despite the elevated odor scores from the AEMs, as
shown in Table 1, the odor score for each individual
AEM unit and test day was never over 3.0 for the test
duration. Furthermore, as given in Table 2, the ammonia
concentration in the airstream outlet from the AEM units
was maintained below 1 ppm. This indicated that acti-
vated carbon exhibited the capability to effectively ad-
sorb NH3 on its surface. The ammonia concentration in
the AEM internal airstreamvaried in the range of 2.61 -
6.30 ppm, as detected by the Dräger sensor inserted into
the gas sampling port. Thus, the exhaust filter exhibited
an ammonia containment efficiency of more than 90%.
The odor removal by the exhaust filter assembled on
AEM units suggests combined mechanics of physisorp-
tion and chemisorption of NH3 for H3PO4 loaded acti-
vated carbon. Ammonia can be absorbed on activated
carbon with pore size similar to its diameter (<4 Å) by
van de Waals forces. Only a small fraction of the acti-
vated carbon surface was utilized during the physisorp-
tion of ammonia due to the fact that activated carbon
hasa larger average pore size (usually 10 - 20 Å). The
impregnation of H3PO4 on the surface of activated carbon
blocks the pores of activated carbon as a result of pore
filling, and thus, creates more micropores that will ex-
hibit higher ammonia removal performance. Another con-
cern about the ammonia adsorption on activated carbon
is that the adsorbed ammonia easily desorbs from the
surface when being purged by air,as a result of the weak
nature of van de Waals forces.
Additionally, the surface acidity of activated carbon is
not ideal for ammonia adsorption. This limited the appli-
cation of activated carbon in ammonia removal as the
key factor that dominates the ammonia adsorption capac-
ity of the activated carbon is the surface chemistry, espe-
cially the acidic groups [8]. The chemical properties of
the carbon surface, especially acidic functional groups,
have a strong influence on NH3 adsorption [17]. The
impregnation of H3PO4 on the activated carbon induces
acidic groups at the basal planets of the activated carbon,
and thus, creates more active sites on the activated car-
bon surface. Ammonia is a basic gas and the introduction
of surface acidity by the impregnation of H3PO4 pro-
motes the NH3 adsorption capacity on activated carbon
via a Brønsted acid-base process [18]. It was reported
that the activated carbon impregnated with H3PO4 has a
highly oxidized surface, contributing to the high surface
acidity (pH < 3) [19]. It was reported that the ammonia
was captured on the surface of the H3PO4-loaded acti-
vated carbon through the following chemical reaction to
form NH4H2PO4 [14]:
H3PO4 + NH3 → NH4H2PO4
Therefore, the creation of acidic functional groups on
the surface of activated carboncaused an increase in the
amount of chemisorbed ammonia as well as an improve-
ment of physisorption properties at low relative pressures.
The deodorization lifetime of the exhaust filter, as in-
dicated by the breakthrough time, is partially dependent
on the amount of H3PO4 impregnated in the adsorbents
[14]. The breakthrough time of the exhaust filter was not
obtained in the current study, but the test results showed
that the new exhaust filter, which uses phosphoric acid
Table 1. Odor evaluation results from the four AEM units with non-ph osphoric acid impregnation exhaust filter.
Odor test scores
Days T + 1 T + 7 T + 14 T + 20 T + 24 T + 30 T + 36 T + 41 T + 45 T + 50
AEM 101 1.05 0 0.06 1.19 1.67 1.21 0.40 0.24 0.88 1.43
AEM 102 1.40 1.24 1.50 1.63 1.44 1.43 1.27 1.18 2.19 2.36
AEM 103 1.60 1.24 0.94 1.69 1.67 1.86 1.80 1.18 1.75 1.93
AEM 104 (c) 1.40 0.47 0.88 0.50 0.67 0.57 1.87 1.82 0.44 0.29
Average 1.35 0.82 0.83 1.12 1.64 1.50 1.64 1.59 1.60 1.90
S.D. 0.28 0.72 0.72 0.28 0.21 0.33 0.33 0.35 0.67 0.46
No. of evaluators 20 16 16 16 15 14 15 17 16 14
Table 2. Ammonia concentrations in the outlet airstreams of the AEM units.
Ammonia concentration in AEM Outlet, ppm
Days T + 21 T + 24 T + 31 T + 36 T + 41 T + 45 T + 50
AEM 101 0.41 0.72 0.75 0.75 0.45 0.45 0.38
AEM 102 <0.20 <0.20 0.26 0.27 <0.20 <0.20 <0.20
AEM 103 <0.20 <0.20 <0.20 <0.20 <0.20 <0.20 <0.20

Z. LU ET AL.
780
impregnated activated carbon as an adsorbent, could ef-
fectively remove ammonia from the exhaust airstream of
mouse habitats for a flight mission of at least 50 days.
4. Conclusion
A high performance, long duration exhaust filter was
developed by NASA Ames Research Center to remove
ammonia from the rodent housing Animal Enclosure
Module, and H3PO4 impregnated activated carbon was
used as its adsorbent. The odor evaluation results sug-
gested that the exhaust filter can effectively control the
odor from the mouse habitats during 45-day (minimum)
operation durations, and maintain the odor from AEM
units within acceptable levels. The AEM exhaust filter
exhibited more than 90% of overall ammonia contain-
ment efficiency. A better understanding of the reaction of
exhaust filter against the odorants in the AEM units
could be achieved through further identification of spe-
cific odor causing components in the exhaust airstream
of the AEM units.
5. Acknowledgements
The current work was supported by the NASA Ames
Research Center Rodent Habitat project.
REFERENCES
[1] M. Ciganek and J. Nega, “Chemical Characterization of
Volatile Organic Compounds on Animal Farms,” Veteri-
narni Medicina, Vol. 53, No. 12, 2008, pp. 641-651.
[2] J. Hartung and V. R. Phillips, “Control of Gases Emis-
sions from Livestock Buildings and Manure Stores,”
Journal of Agricultural Engineering Research, Vol. 57,
No. 3, 1994, pp. 173-189.
http://dx.doi.org/10.1006/jaer.1994.1017
[3] D. H. O’Neill and V. R. Phillips, “A Review of the Con-
trol of Odour Nuisance from Livestock Buildings: Part 3,
Properties of the Odorous Substances which Have Been
Identified in Livestock Wastes or in the Air around
Them,” Journal of Agricultural Engineering Research,
Vol. 53, 1992, pp. 23-50.
http://dx.doi.org/10.1016/0021-8634(92)80072-Z
[4] S. S. Schiffman, J. L. Bennett and J. H. Raymer, “Quanti-
fication of Odors and Odorants from Swine Operations in
North Carolina,” Agricultural and Forest Meteorology,
Vol. 108, No. 3, 2001, pp. 213-240.
http://dx.doi.org/10.1016/S0168-1923(01)00239-8
[5] L. D. Jacobson, J. R. Bicudo, D. R. Schmidt, S. Wood-
Gay, R. S. Gates and S. J. Hoff, “Air Emissions from
Animal Production Buildings,” ISAH, Mexico, 2003.
[6] ACGIH, “Threshold Limit Values for Chemical Sub-
stances and Physical Agents and Biological Exposure In-
dices,” American Conference of Governmental Industrial
Hygienists, Cincinnati, 1992.
[7] Spacecraft Maximum Allowable Concentrations for Air-
borne Contaminants, NASA JSC, 1999.
[8] M. Goncalves, L. Sánchez-Garcia, E. de Oliveira Jardim,
J. Silvestre-Albero and F. Rodriguez-Reinoso, “Ammonia
Removal Using Activated Carbon: Effect of the Surface
Chemistry in Dry and Moist Conditions,” Environmental
Science & Technology, Vol. 45, No. 24, 2011, pp. 10605-
10610. http://dx.doi.org/10.1021/es203093v
[9] A. E. Ghaly and K. N. MacDonald, “Development and
Testing of an Ammonia Removal Unit From the Exhaust
Gas of a Manure Drying System,” American Journal of
Environmental Science, Vol. 9, No. 1, 2013, pp. 51-61.
http://dx.doi.org/10.3844/ajessp.2013.51.61
[10] F. Stoeckli, A. Guillot and A. M. Slasli, “Specific and
Non-Specific Interaction between Ammonia and Acti-
vated Carbon,” Carbon, Vol. 42, No. 8-9, 2004, pp. 1619-
1624. http://dx.doi.org/10.1016/j.carbon.2004.02.034
[11] T. Bandosz and C. Petit, “On the Reactive Adsorption of
Ammonia on Activated Carbon Modified by Impregna-
tion with Inorganic Compounds,” Journal of Colloid and
Interface Science, Vol. 338, No. 2, 2009, pp. 329-345.
http://dx.doi.org/10.1016/j.jcis.2009.06.039
[12] C. Petit, C. Karwacki, G. Peterson and T. Bandosz, “In-
teractions of Ammonia with the Surface of Microporous
Carbon Impregnated with Transition Metal Chlorides,”
Journal of Physical Chemistry C, Vol. 111, No. 34, 2007,
pp. 12705-12714. http://dx.doi.org/10.1021/jp072066n
[13] L. M. Le Leuch and T. J. Bandosz, “The Role of Water
and Surface Acidity on the Reactive Adsorption of Am-
monia on Modified Activated Carbon,” Carbon, Vol. 45,
No. 3, 2007, pp. 568-578.
http://dx.doi.org/10.1016/j.carbon.2006.10.016
[14] A. Oya and W. Iu, “Deodorization Performance of Char-
coal Particles Loaded with Orthophosphoric Acid against
Ammonia and Trimethylamine,” Carbon, Vol. 40, No. 9,
2002, pp. 1391-1399.
http://dx.doi.org/10.1016/S0008-6223(01)00273-1
[15] K. Maruyama, H. Takagi, M. Kodama, H. Hatori, Y. Ya-
mada, R. Asakura, H. Izumida and M. Mitsuhiro, “Am-
monia Adsorption on Porous Carbon with Acidic Func-
tional Groups,” TANSO, Vol. 203, No. 208. 2003, pp.
109-113. http://dx.doi.org/10.7209/tanso.2003.109
[16] C. Muller, “Comparison of Chemical Filters for the Con-
trol of Airborne Molecular Contamination,” Journal of
the IEST, Vol. 50, No. 2, 2007, pp. 52-72.
[17] J. Guo, W. Xu, Y, Chen and A. Lu, “Adsorption of NH3
onto Activated Carbon Prepared from Palm Shells Im-
pregnated with H2SO4,” Journal of Colloid and Interface
Science, Vol. 281, No. 3, 2005, pp. 285-290.
http://dx.doi.org/10.1016/j.jcis.2004.08.101
[18] C. Huang, H. Li and C. Chen, “Effect of Surface Acidic
Oxides of Activated Carbon on Adsorption of Ammonia,”
Journal of Hazardous Materials, Vol. 311, 2008, pp. 311-
314.
[19] J. Laine, A. Calafat and M. Labady, “Preparation and
Characterization of Activated Carbon from Coconet Shell
Impregnated with Phosphoric Acid,” Carbon, Vol. 27, No.
2, 1989, pp. 191-195.
http://dx.doi.org/10.1016/0008-6223(89)90123-1
Open Access AJAC