TITLE:  Fire-Trol Holdings, LLC, B-293103; B-293254, February 2, 2004
BNUMBER:  B-293103; B-293254
DATE:  February 2, 2004
**********************************************************************
Fire-Trol Holdings, LLC, B-293103; B-293254, February 2, 2004

   DOCUMENT FOR PUBLIC RELEASE                                                
The decision issued on the date below was subject to a GAO Protective      
Order.  This version has been approved for public release.                 

   Decision
    
Matter of:   Fire-Trol Holdings, LLC
    
File:            B-293103; B-293254
    
Date:              February 2, 2004
    
Paul F. Dauer, Esq., and Jennifer L. McCready, Esq., Best Best & Krieger,
for the protester.
John G. Horan, Esq., and Jason A. Carey, Esq., McDermott, Will & Emery,
for Astaris, LLC, an intervenor.
Marion T. Cordova, Esq., Department of Agriculture, for the agency.
Edward T. Goldstein, Esq., and Christine S. Melody, Esq., Office of the
General Counsel, GAO, participated in the preparation of the decision.
DIGEST
    
1.  Protest that agency improperly included competitor*s products on
qualified products lists (QPL) is denied where agency reasonably
determined that re-testing of modified or changed products was unnecessary
before listing those products on the QPL and where the agency reasonably
determined the products complied with its qualification requirements.
    
2.  Agency*s inclusion of patent indemnity clause in solicitations is
unobjectionable where the solicitations are for supplies that are sold or
offered for sale to the public in the commercial market.
DECISION
    
Fire-Trol Holdings, LLC protests the inclusion of a competitor*s products
on the qualified products lists (QPL) under invitation for bids (IFB) Nos.
49-03-10, 49-03-15, and 49-03-16 and under request for proposals (RFP)
Nos. 49-03-09, 49-03-13, and 49‑03-14, issued by the United States
Department of Agriculture (USDA), Forest Service. Fire-Trol contends that
the protested products, manufactured by Astaris, LLC, fail to satisfy
several of the specifications required for approval. Fire-Trol also
challenges the agency*s inclusion of a patent indemnity clause in the
solicitations.
    

   We deny the protests.
Background
    
The Forest Service purchases chemical-based, water-soluble, long-term fire
retardants for use in preventing and suppressing *wildland* fires
throughout the United States.  The retardants at issue in this case are
those that are dropped from the air by fixed-wing aircraft (that is,
aircraft other than helicopters).  Generally speaking, a retardant product
has four main components: (1) its salt content, which largely determines
the product*s combustion-retarding effectiveness; (2) its colorant; (3)
its corrosion inhibitor; and (4) its thickening agent.
    
The fire retardant products used by the Forest Service are manufactured
and supplied in various forms with different properties.  Some retardant
products are supplied in a powder form (dry concentrate), which is mixed
with water prior to use.  Others are supplied in a liquid form (wet
concentrate), also referred to as liquid concentrate products, and are
diluted with water prior to use.  A given liquid concentrate product may
be available in different concentrations; however, once diluted, they
*should be identical in concentration and performance.*[1]  Agency Report
(AR) Tab AO, Agency Memorandum of Law at 4. 
    
Fire retardant products also have varying color characteristics.  Some are
uncolored, while others are classified as colored due to the fact that
they contain iron oxide.  Generally, retardant products that contain iron
oxide have an *R* designation since iron oxide imbues the product with a
reddish color.  Retardants with an *F* designation are those that do not
contain iron oxide but have what is referred to as a fugitive color. 
Fugitive products are visible for a short period of time but quickly fade
into the landscape. 
    
In addition, products have varying degrees of viscosity.  *Unthickened*
products (essentially water-like) have a lower viscosity than *thickened*
products, which contain some form of thickening agent such as gum and are
generally categorized as of high, medium, or low viscosity, depending upon
the amount of thickening agent in the product.  The more thickening agent
a product contains, the higher its viscosity.
    
The IFBs, which were issued on October 21, 2003, anticipate the award of
fixed-price indefinite-quantity requirements contracts for the supply of
long-term fire retardant at permanent firefighting bases throughout the
United States as well as the attendant services of mixing and storing the
retardant and loading of the retardant into airtankers (or as otherwise
directed by the agency).[2]  The RFPs, on the other hand, which were
issued on October 31, 2003, anticipate the award of fixed-price
indefinite-quantity requirements contracts, which are solely for the
supply of
long-term fire retardant at various permanent firefighting bases; they are
essentially *bulk* purchase contracts for fire retardant.[3]
    
All of the solicitations provide that the retardants must be qualified and
approved in accordance with the governing Forest Service specification,
5100-304b, dated January 2000.  Products that have been approved in
accordance with specification 5100-304b are listed on the Forest Service*s
fire retardant QPL and can be offered without further qualification. 
    
With regard to qualification testing, Forest Service specification
5100-304b states as follows:
    
4.3.1.  Qualification Tests.  The samples submitted shall be subjected to
the applicable tests listed in 4.5 to determine if they meet the
requirements of 3.4 through 3.13.  These tests will be conducted at Forest
Service facilities or other laboratories designated by the Forest
Service.  Following satisfactory completion of laboratory testing, an
operational field evaluation may be conducted as required in 3.14.
    
Section 4.5 of the specification 5100-304b, referenced in section 4.3.1,
sets forth a range of qualification tests that a retardant must satisfy in
order to be listed on the QPL as fully qualified.  As it relates to this
case, the specification includes the following testing requirements:  (1)
combustion-retarding effectiveness (section 4.5.2); (2) corrosion (section
4.5.6); (3) product stability (section 4.5.5); (4) air-drop
characteristics (section 4.5.9); (5) field visibility (section 4.5.10);
and (6) operational field evaluation testing (section 4.5.11).[4]  Some of
the above-identified tests are considered *lab* tests because they are
performed in a laboratory under controlled conditions.  These would
include the tests for combustion-retarding effectiveness, corrosion, and
product stability.  Testing for air-drop characteristics, field
visibility, and the operational field evaluation tests are conducted in
the field under variable conditions.
    
The specification further describes qualification of changed or modified
products in section 4.3.4:
    
4.3.4.  Qualification of Changed or Modified Product.  Prior to making any
changes in the formulation, including the type, quantity, quality,
processing, supplier, manufacturer, or manufacturing site of individual
ingredients, the retardant manufacturer shall notify the National
Director, Fire and Aviation Management, USDA Forest Service.  At the
option of the Forest Service, qualification testing may be required.
    
According to the agency, the submission process for qualification testing
begins with a manufacturer submitting a letter to the Forest Service
requesting that it evaluate one of its products.  Before testing is
conducted, the manufacturer provides necessary information pertaining to
its product, referred to as the *product information sheet.*  The
information sheet includes the chemical formula, or *recipe* for the
product, which includes the manufacturer of each ingredient, the quality
of each ingredient, and the amount of each ingredient by percentage weight
in concentrate and at the use-level mix of the product.  The manufacturer
also assigns the product a *lot number,* which is unique for that specific
product.  If, for example, there is a change to the manufacturer of an
ingredient, that product would require a separate lot number.  See AR, Tab
AQ, Statement by Project Leader for the Wildland Fire Chemical Systems
Program at the Forest Service*s Missoula Technology and Development
Center, Nov. 24, 2003.
    
The agency notes that it does not perform each of the above-described
tests for every product submitted for qualification.  In fact, the agency
indicates that none of the products on the QPL, including the protester*s
products, have been subjected to the full range of tests identified in the
solicitation.  See Hearing Transcript (Tr.)
at 202-03.[5]  In this regard, while, according to the agency, corrosion
and stability testing are always performed, with respect to other tests,
the agency determines, on a case-by-case basis, whether a particular test
is necessary based on its understanding of the product*s ingredients and
its probable test results. 
    
Thus, the agency explains that it looks to the salt content of the product
to determine whether burn testing is required.  Where the salt content is
sufficiently similar to another product submitted by the manufacturer and
that product has been successfully burn tested, the agency may determine
that the product meets the agency*s combustion-retarding effectiveness
requirements, without burn testing the product.  Evidently, over the
years, the agency has applied this method of evaluating a product*s
combustion-retarding effectiveness with respect to products from both the
protester and the intervenor.  See, e.g., AR, Tab R, Agency Letter to
Fire-Trol, Apr. 15, 2002 (agreeing with the protester that certain of its
products submitted for qualification did not require burn testing based on
the fact that they contained more active salts than another one of the
protester*s products, which had been successfully burn tested).   
    
With regard to air-drop testing, the agency explains that it has not
performed this test in years because it has sufficient data indicating
that products with a particular thickening agent (e.g., gum-thickened
products) have very similar air-drop characteristics.  See Tr. at 45. 
According to the agency, scheduling this test is challenging:  it can only
be performed in the late fall or early spring (since it must be performed
under certain weather conditions), when an airtanker is available, and
when there is a location available to safely drop retardant and perform
measurements.  See AR, Tab AQ, supra.  Moreover, the agency notes that the
test is time-consuming as well as very expensive, with costs ranging from
tens to hundreds of thousands of dollars.  Id.  Accordingly, the Forest
Service will not evaluate a product*s air-drop characteristics, through
actual testing, unless the product uses a new type of thickening agent. 
Id.   
    
Similarly, the agency indicates that for the last several years, it has
not performed visibility testing on products submitted for qualification
that use iron oxide for its colorant.  See Tr. at 45.  Because visibility
testing is time-consuming, expensive, and highly subjective, the agency
explains, as long as a product submitted for qualification contains iron
oxide in an amount that has been used for numerous years, there is no need
for the product to be visibility tested as part of the qualification
process.  See AR, Tab AQ, supra; Tr. at 50.       
    
Specific Products
    
The solicitations at issue in the protests include the agency*s QPL of
September 5, 2003, as an attachment.  This QPL listed the following
Astaris fire retardant products as *fully qualified*:  (1) Phos-Check HV-R
(3.6:1 mix ratio); (2) Phos-Check HV-R (3.1:1 mix ratio); (3) Phos-Check
MV-R; (4) Phos Check LV-R; (5) Phos-Check
LC-95A; and (6) Phos-Check LC-95D.    
    
All of the products at issue were subjected to laboratory testing for
corrosion and stability and were determined to be acceptable.[6]  None of
the products, however, when submitted for qualification, were subjected to
burn testing, visibility testing, or testing for air-drop characteristics.
With regard to air-drop testing, since the six products are all
gum-thickened products, the agency determined that testing their air-drop
characteristics was unnecessary because, as explained above, gum-thickened
products, according to the agency, have proven to have similar and
satisfactory air-drop characteristics.  The agency determined that
visibility testing was not required because all of the products use iron
oxide for their colorant, and, as described above, the agency does not
test the visibility of products that use iron oxide.     
    
As to burn testing, the agency determined that testing of the HV-R, MV-R,
and LV-R products was not necessary because they all contained higher
retardant salt contents than a similar Astaris product, D75-F, which was
successfully burn tested and qualified.  See AR, Tab I, Revising Status of
Fluid Concentrate Products Having a 3.6:1 or 3.7:1 Mix Ratio, June 2,
2003.  According to the agency, the higher the salt content of a similar
product the better its combustion-retarding effectiveness.  Each of these
products uses a combination of retardant salts to provide its fire
retarding characteristics.  According to the agency, while the amounts of
the various salts may not be exactly the same, they can all be
standardized by calculating their P2O5 (fire retarding salt) equivalent. 
Based on the P2O5 value, the agency indicates that it can calculate the
combustion-retarding characteristics of a product based on results from
past tests.  Since HV-R, MV-R and LV-R all have higher calculated P2O5
values than D-75R, the agency determined that these products did not
require burn testing. 
    
Similarly, as to LC-95A and LC-95D, the agency determined that burn
testing was not necessary because they contained higher salt contents than
similar Astaris products that had been successfully burn tested.  The
agency explains that LC-95A and LC-95D are wet concentrates that are
formulated using a liquid ammonium polyphosphate (APP) fertilizer as the
source of their P2O5.  The agency also notes that the burn characteristics
of APP can vary depending on its source; thus, knowing the source of the
APP is critical when evaluating a product*s combustion-retarding
properties.  In this case, Simplot 11-37-0 is the source of the APP for
LC-95A and Simplot 10-34-0 is the source of APP for LC-95D. 
    
With regard to LC-95A, the agency determined that burn testing was
unnecessary because it had a higher salt content than another similar
Astaris wet concentrate product, lot number 1051695-E, which had been
successfully burn tested.  Both
LC-95A and lot number 1051695-E use Simplot 11-37-0 as the source of their
fire retarding salt.  Burn testing for LC-95D was determined to be
unnecessary because it had a higher salt content than lot number 1051615-E
and LRA, both of which were successfully burn tested.  LC-95D, lot number
105615-E and LRA all use Simplot
10-34-0 as the source of their fire retarding salt.      
    
As to operational field evaluation testing, HV-R (3.1:1) was subjected to
a full operational field evaluation and LV-R was subjected to a
monitor-level field evaluation.  See AR, Tab B, Letter, Subject: Status of
Field Testing of Phos-Check HV-R, MV-R, and LV-R, Jan. 12, 1994; Tab C,
Letter, Subject: Results of the Laboratory Evaluation of the Corrosivity
of Field Test Samples of Phos-Check HV-R, Feb. 10, 1994.  Since HV-R
(3.1:1) had passed the full field evaluation, the agency determined that a
field evaluation for HV-R (3.6:1) was not necessary.  The agency also
determined that a field evaluation was not necessary for MV-R since LV-R
had successfully passed the monitor-level field evaluation. 
    
With regard to LC-95A and LC-95D, the agency determined that field testing
was not necessary because they were similar to another Astaris
formulation, LRA, which had been successfully field tested.  According to
the agency, the LC-95 formulations were the same as, or contained
improvements to, the LRA formulation.  Specifically, the amount of iron
oxide was the same for these products, thus leading the agency to conclude
that they should have similar visibility characteristics in the field.  In
addition, the agency considered the LC-95 formulations to be an
enhancement of LRA because they are gum-thickened, improving their
air-drop performance when compared with LRA, an unthickened product.  The
LC-95 formulations also use a different corrosion inhibitor, which had
*very good* results in the laboratory corrosion tests and, therefore, did
not trigger the need for a field test according to the agency.  Tr. at
193.           
    
In its protest, Fire-Trol  contends that all of the above six identified
Astaris products were improperly listed as fully qualified on the
solicitations* QPL.  As a threshold matter, the protester contends that,
under the terms of the specification, the agency did not have the
authority to waive burn testing and visibility testing for the products. 
Alternatively, to the extent the agency did have such authority, the
protester maintains that it abused its discretion in determining that
LV-R, MV-R, and the LC-95 formulations did not require air-drop testing. 
In addition, the protester asserts that both HV-R formulations should not
have been qualified because HV-R (3.1:1) has demonstrated problems with
corrosion when used in the field.  The protester also asserts that the
agency abused its discretion in determining that operational field testing
was not necessary for MV-R and the LC-95 formulations, and challenges the
scope and results of the monitor-level field evaluation of LV-R.
    
Where a solicitation requires that a product be qualified or approved, it
is improper to include in the list of approved products an item that has
not been properly approved and does not satisfy the applicable
specifications; an agency*s including such a product in the list may
constitute a basis for sustaining a protest if the agency*s action
prejudices another offeror.  Chemonics Indus., Inc., B-260284, Apr. 19,
1995, 95-1 CPD P: 206 at 3.  The conduct of qualification testing,
however, is an area where contracting agencies have broad discretion so
long as their actions are reasonable and do not prejudice potential
offerors, by, for example, treating them unequally.  Id. at 4.
    
Discretion to Waive Testing
    
In support of its position that the agency could not waive burn or
visibility testing for the subject products, the protester points to
language in the specification stating that the combustion-retarding
effectiveness of a product *shall be determined by burning treated 8 foot
fuel beds* and that a product *shall be tested to determine visibility on
a variety of fuel types and conditions (slope, aspect, daylight
conditions, and weather).*  Specification 5100-304b S:S: 4.5.2, 4.5.10
(emphasis added). 
    
The agency contends, however, that it had the authority under the terms of
the specification not to conduct burn testing or visibility testing for
the products at issue.  In support of its position, the agency points to
the section of the specification pertaining to qualification of changed or
modified products, which provides:  *At the option of the Forest Service,
qualification testing may be required.*  Specification 5100-304b S: 4.3.4
(emphasis added).  According to the agency, the products at issue are all
changed or modified versions of products that had already been qualified,
and thus re-testing of these changed or modified products was within its
discretion.  The agency also relies on language in the specification
regarding qualification testing in general, which states: *The samples
submitted shall be subjected to the applicable tests listed in 4.5 to
determine if they meet the requirements of 3.4 through 3.13.* 
Specification 5100-304b S: 4.3.1 (emphasis added).  The agency argues that
use of the term *applicable* means that not all of the tests listed in
section 4.5 of the specification will necessarily be conducted and that
the agency has the discretion to determine which tests are *applicable.*

    
The protester contends that the agency*s interpretation is unreasonable. 
According to the protester, none of the products at issue are
modifications of prior approved products, and thus section 4.3.4 of the
specification, on which the agency relies, does not apply.  Further,
Fire-Trol argues, even if section 4.3.4 did apply, because the agency
determined that some qualification tests were necessary, the agency was
required to conduct the full battery of tests identified in the
specification, which includes burn and visibility testing.  Moreover, the
protester argues that the phrase *applicable tests* in section 4.3.1 does
not mean that all tests are discretionary, but refers only to those areas
where the specification expressly affords the agency discretion in
deciding whether to conduct a test (specifically, air-drop and operational
field evaluation testing).[7]  Since the agency used clear language to
retain the discretion to waive air-drop testing and operational field
evaluation testing, the protester adds that the agency obviously intended
not to provide for such discretion where it did not include similar
language, i.e., with regard to burn and visibility testing.   
    
As an initial matter, it should be noted that while Fire-Trol now contends
that the language of the solicitation cannot be read to allow for the
waiving of burn testing and visibility testing based on the
characteristics and performance of other similar products, Fire-Trol had
actively sought and obtained such waivers with regard to its products in
the past.  Thus, the protester*s argument that the agency*s interpretation
of its specification, one that the protester had previously taken
advantage of, is unreasonable and improper, appears disingenuous and
unpersuasive on its face.  In any event, the protester*s narrow
interpretation of the specification terms, and how they were applied to
the six products at issue, is unreasonable in this case. 
    
First, contrary to the protester*s assertion, when evaluating a modified
or changed product, the specification does not expressly require that the
agency conduct the full range of qualification testing set forth in the
specification; rather, the specification simply states that *qualification
testing may be required.*  Specification 5100-304b
S: 4.3.4.  The agency has, as a matter of practice, one that the protester
was fully aware of and benefited from, determined that the scope of the
qualification testing of a changed or modified product depends on the
nature of the changes or modifications.  Given the broad discretion
afforded agencies with regard to the conduct of qualification testing, and
the discretion afforded the agency by the specification itself in
determining whether qualification testing of a modified or changed product
is required at all, the agency has reasonably interpreted specification
section 4.3.4 as providing it with the latitude to only conduct those
qualification tests that it deems necessary for changed or modified
products based on the nature of the changes or modifications.[8]
    
Second, it is for the agency to determine, based on its understanding of
the retardants and the specification requirements, whether a product
submitted for qualification is a changed or modified version.  There is
nothing in the record to suggest that the agency*s decisions in this
regard were unreasonable.  As the agency noted, when determining whether
to test the HV-R, MV-R, and LV-R family of products, it compared the mixed
form of these products with the mixed form of a *similar* Astaris product,
D75-F, which had been successfully burn tested, qualified, and used since
1984.  Tr. at 169.  According to the agency, the composition of the mixed
form of these products was essentially the same as the composition of the
D75-F product, with the main difference being the colorant used in D75-F
(HV-R,
MV-R, and LV-R all contain iron oxide for their colorant while D75-F is a
fugitive colored product).  Moreover, the agency*s determination that the
mixed versions of these products were all similar is supported by the
information set forth in the confidential disclosure sheets for these
products, which contain their formulations.  These sheets reflect that all
of the products have the same ingredients, other than colorant, and the
ingredients vary only slightly in amount.  Thus, it was not unreasonable
for the agency to conclude that these products were merely different
versions of the same product. 
    
Similarly, with regard to LC-95A and LC-95D, the agency equated these
products to other Astaris ammonium polyphosphate formulations that had
undergone various qualification tests.  For example, LC-95A was determined
to be *similar* to lot number 1051695-E, and LC-95D was determined to be
similar to Phos-Check LRA.  See AR, Tab AH, Letter, Subject: Analysis to
Determine Need for Operational Field Evaluation of Phos-Check LC-95A and
Phos-Check LC-95D. 
    
The main difference between the products is the corrosion inhibitor used
and the fact that LC-95A and LC-95D were thickened products to improve
their air-drop characteristics.  Again, the confidential disclosure sheets
for these products reveal that their formulations are very similar.  Thus,
it was not unreasonable for the agency to conclude that these products
were, in essence, merely different versions or variations of the same
product.
    
    
    
Burn Testing
    
The protester alternatively argues that even if the agency had the
discretion not to burn test the products at issue, it abused that
discretion because changes to minor ingredients, such as the addition of
or change in thickener, can affect the combustion-retarding effectiveness
of a product.  In support of its argument, the protester relies primarily
on a Forest Service research article stating *retardant effectiveness
cannot be accurately predicted by analysis of total concentration of
active salts . . . .*  Protest, Tab J, Flammability Reduction Comparisons
of Four Forest Fire Retardants, April 1988, Aylmer D. Blakely, at 1.  The
protester also generally argues that it was an abuse of discretion to rely
on the burn test results of another product because the multiple burn test
results for a particular product can be variable.[9]
    
The protester*s challenges to the agency*s conclusions regarding the
combustion-retarding effectiveness of the products at issue, and its
concerns about the inherent unreliability of burn testing in general,
amount to little more than speculation and disagreement concerning the
agency*s burn testing procedures and fundamental position that, because
the products at issue all had higher active salt content than products
which had been successfully burn tested, it could reliably conclude that
burn testing was not necessary.[10]  While the report cited by the
protester suggests that *some* additives to a product such as its
thickener or corrosion inhibitor *can* affect a product*s fire retarding
characteristics, the protester fails to demonstrate how any of the changes
to the products at issue would have negatively affected their burn
characteristics, particularly where they had higher active salt than the
burn tested products.  Given the agency*s discretion in determining
whether a product has met its specification requirements, and the agency*s
reasoned approach to analyzing the burn characteristics of the products at
issue, we find the protester*s challenges unpersuasive.
    
Visibility Testing
    
In terms of visibility testing, the protester argues that the agency
abused its discretion by waiving testing for MV-R, LV-R, LC-95A and
LC-95D.  According to the agency, visibility testing was not required for
these products because they all contain iron oxide, and products
containing iron oxide have proven over the years to be sufficiently
visible when used in the field.  Moreover, the agency noted that MV-R and
LV-R contain the same amount of iron oxide as D75-R, which had been fully
qualified and successfully used in the field and that LC-95A and LC-95D
contained the same amount of iron oxide as LRA, which was also determined
to have met the agency*s visibility requirements. 
    
The protester maintains, at least with respect to MV-R and LV-R, that
viscosity also affects a product*s visibility and that the agency failed
to take this into account when determining that visibility testing for
these products was unnecessary.  As support for its contention that
viscosity must be considered when evaluating visibility, the protester
highlights the fact that during operational field testing of fugitive
colored products, a high viscosity product was determined to have
satisfactory visibility characteristics, while medium and low viscosity
products were found to have insufficient visibility.  AR, Tab B, supra. 
Based on these field test experiences, the protester draws the more
general inference that viscosity must be an important factor when
considering visibility for any product, not just fugitive products.  The
protester, however, clearly takes this inference further than the record
supports.  Because fugitive products have inherently different visibility
characteristics than iron oxide colored products, the protester*s tenuous
conclusions do not provide a basis for questioning the agency*s judgment
that it is not necessary to test the visibility characteristics of iron
oxide products.
    
Corrosion Testing
    
Fire-Trol also argues that HV-R (3.1:1) and (3.6:1), MV-R, and LV-R should
not have been qualified because there is evidence of HV-R having caused
excessive corrosion of storage tanks when used in the field.[11]  This
challenge, however, is flawed in several respects.  As an initial matter,
it should be noted that each of the products challenged in this regard
successfully passed the agency*s laboratory corrosion tests, and the
laboratory tests are the only corrosion qualification requirements under
specification 5100-304b.  Moreover, the agency performed corrosion testing
on field samples of the HV-R products and these samples were found
acceptable.  The agency reasonably states that it does not make judgments
about products based on field observations of corrosion in storage tanks
because such conclusions would be inherently unreliable since, before a
product is used, there is no way to know the tank*s condition.  Tr.
59-60.  The agency notes that a storage tank*s condition depends on a
variety of factors, including where a tank is procured, its age,
maintenance record, and what other products have been used in the tank in
the past.[12]  Tr. at 59-61.  Given that these products all passed the
agency*s laboratory corrosion testing and in some instances corrosion
testing of field samples, the unreliable anecdotal evidence concerning
corrosion found in field tanks after they had be used with HV-R, does not
suggest that the agency acted in an arbitrary or unreasonable manner in
qualifying HV-R, MV-R, or LV-R.
    
In addition, the protester argues that by qualifying these products the
agency disregarded the fact that there were problems with corrosion test
results performed on field samples.  According to the protester, given the
documented problems with these products, they should have been subjected
to more extensive field testing for corrosion before being qualified. 
While it is true that samples of MV-R and LV-R taken from the field did
not pass corrosion tests in 1991, the Forest Service determined that this
was due to the fact that they were not being adequately recirculated, thus
compromising the samples tested.  Astaris and the agency, however,
addressed the recirculation issue, and ultimately corrosion tests of field
samples of HV-R (3.1:1) proved satisfactory, leading to the qualification
of HV-R by 1994.  See AR, Tab C, supra.  Given the positive results with
HV-R, which, in its mixed form, is essentially the same as MV-R and LV-R,
it was reasonable for the agency to determine that further field testing
of these products for corrosion was unnecessary.[13]  See AR, Tab I,
supra.
Air-drop Testing
The agency states that it did not air-drop test any of the products at
issue because they are all gum-thickened and all gum-thickened products
are considered sufficient.  While recognizing that air-drop testing *is
discretionary with respect to any particular retardant,* the protester
maintains that the agency should have air-drop tested MV-R because the
agency did not have enough experience with the aerial delivery of this
product, and it should have tested LC-95A and LC-95D because the agency
does not have any experience with similar gum-thickened products. 
Protester*s Response to Documents, Astaris Comments, and Supplemental
Agency Report, Dec. 31, 2003, at 21, 28-29.  We view the protester*s
challenges in this regard as mere disagreement with the agency*s
conclusions, which is insufficient to render those conclusions
unreasonable.
Operational Field Evaluation Testing
With regard to LV-R, Fire-Trol argues that this product was improperly
qualified because it failed to demonstrate its ability to maintain
adequate viscosity when tested during the course of a monitored
operational field evaluation.[14]  The agency recognized that in some
instances the test results for LV-R samples revealed very low viscosity
levels; however, upon further inquiry, the agency determined that these
low levels were not due to problems with the product.  Tr. at 366.  In
some instances too much water had been mistakenly mixed with the product,
decreasing its viscosity.  Tr. at 362.  The agency also determined that in
some instances the product had been mixed with what it believed to be
contaminated water (water from untreated lakes or ponds), which can
degrade the gum-thickener and decrease viscosity. 
Tr. at 215-16.  Moreover, the record reflects that in many instances the
samples tested had satisfactory viscosity levels.  See AR, Tab DV,
Operational Field Evaluation Results.  Given this record, there is nothing
to suggest that the agency abused its discretion in qualifying LV-R based
on the results of its monitored operational field test.
Turning to LC-95A and LC-95D, the protester asserts that these products
should have been subjected to operational field testing.  Field testing of
these products was not required because they were determined to be similar
to LRA, which had been successfully field tested.  The protester
maintains, however, that unlike LRA, LC-95A and LC-95D were gum-thickened
and that they had a different corrosion inhibitor than LRA.  The agency
specifically considered these differences and determined that field
testing was not required.[15]  AR, Tab AH, supra.  Again, the protester
merely presents its disagreement with the agency*s considered judgment
that further testing was not necessary for these products, and such
disagreement will not serve as a basis for us to question the agency*s
actions.
Inclusion of Patent Indemnity Clause
As a final matter, Fire-Trol challenges the agency*s inclusion of a patent
indemnity clause  (FAR S: 52.227-3), arguing that it was improperly
included in the solicitations since the products are not sold to the
general public in the commercial market.  The clause at issue protects the
government from liability for patent infringement by requiring a
contractor to reimburse the government for liability for infringement
resulting from performance of the contract.  FAR S: 27.203-1.  With
respect to the circumstances under which the provision may not be used,
FAR S: 27.203-1 expressly states as follows (emphasis added):
(b) A patent indemnity clause shall not be used in the following
situations: . . .
(2) When the contract is for supplies or services (or such items with
relatively minor modifications) that clearly are not or have not been sold
or offered for sale by any supplier to the public in the commercial open
market.
Given the use of the term *clearly,* the FAR sets a high threshold for
determining that a patent indemnification clause cannot be used.  Thus, if
it is not clear that the government is purchasing a product that has not
been sold or offered for sale to the public in the commercial market, the
prohibition against the use of FAR
S: 52.227-3 would not apply.[16]
In this case, the agency acted reasonably by including the FAR provision. 
As an initial matter, the agency highlights the fact that Astaris sells
the very products at issue in this case to a commercial company, Hunot
Retardant Company.[17]  See Contracting Officer*s Statement at 1.  The
agency also notes that the solicitation is for the purchase of
chemical-based, water-soluble fire retardants and cites various
advertisements for such retardants that are sold for use on homes or with
live trees.  One such advertisement states, **Protect your home with the
same powerful fire retardant used by the U.S. Forest Service.**  Agency
Response to Protester*s Comments at 9.  Moreover, the intervenor notes
that it has sold its fire retardant products to timber companies in Chile
as well as the United States, and adds that Fire-Trol has sold its fire
retardant products to timber companies in Chile as well.  See Decl. of
Business Director, Fire Safety, for Astaris, Jan. 5, 2003.  While the
protester contends that in many cases the cited products are different
from the products contemplated under the solicitation because they are not
for use from fixed-wing aircraft (noting that they therefore would not be
gum-thickened or contain iron oxide or corrosion inhibitor) or because
they are not listed on the QPL, the agency emphasizes the fact that the
cited products are all chemical-based,
water-soluble fire retardants.[18]  Because the record reflects that the
solicitations are for the procurement of products that have been sold to
or offered for sale to commercial entities in arm*s-length commercial
transactions, the agency did not act unreasonably by including the patent
indemnification clause in the subject solicitations.
    
The protests are denied.
    
Anthony H. Gamboa
General Counsel
    
    

   ------------------------

   [1] For example, a liquid concentrate product with a water to product mix
ratio of 3.1:1 requires the dilution of 3.1 parts water to one part
product, while a more concentrated version of the same product (one with
less water to start with), for instance a version with a mix ratio of
3.6:1, would require the dilution of 3.6 parts water to one part product. 
Both mix ratios of the retardant product when diluted for use in an
aircraft, however, yield the same concentration.
[2] IFB No. 49-03-10 is for a single base year, with no option years; IFB
No. 49-03-15 is for a single base year, with 1 option year; and IFB No.
49-03-16 is for a single base year, with 2 option years.
[3] IFB No. 49-03-9 is for a single base year, with no option years; IFB
No. 49-03-13 is for a single base year, with 1 option year; and IFB No.
49-03-14 is for a single base year, with 2 option years.
[4] The other specification testing requirements, which are not at issue
in this case, are: (1) mammalian toxicity and irritation; (2) optimum
mixing; (3) physical properties (including active salt content, viscosity,
steady state viscosity, density, pH value, and refractometer reading); (4)
pumpability; and (5) abrasion.
[5] Our Office conducted a hearing on the protest during which testimony
was received from the Program Leader for the Wildland Fire Chemical
Systems Program at the Forest Service*s Missoula Technology and
Development Center, as well as the Project Leader for Fire Chemicals
within the Forest Service*s Wildland Fire Chemical Systems Program.
[6] During operational field evaluation testing of LV-R and MV-R, which
began in 1991, field samples of these products were subjected to
laboratory corrosion testing.  Based on the results of those tests, the
products were not qualified.  See AR, Tab B, Letter, Subject: Status of
Field Testing of Phos-Check HV-R, MV-R, and LV-R, Jan. 12, 1994, at 2; Tab
BR, Letter, Subject: 1994 Operational Field Testing of Phos-Check
retardants, Jan. 14, 1994, at 2; and Tab DL, Status Report for the Fire
Chemicals Executive Committee, Fluid and Liquid Concentrates, Nov. 27,
1992, at 11-12.  According to the agency, these corrosion test results
were ultimately discounted due to non-representative product samples being
used for the corrosion testing, which resulted from problems with
recirculation of the product during storage, problems that were later
remedied.  See Tr. 207-09; cf. AR, Tab AR, Operational Evaluation Test
Plan for Phos-Check Fluid Concentrate Fire Retardants HV-R and HV-F, May
1992, at 1 (discussing recirculation problems with LV-R and MV-R during
1991 field evaluation of these products).  
[7] Specifically, section 4.5.9 of the specification provides for testing
a retardant*s
air-drop characteristics *as deemed necessary by the Forest Service . . .
.*  With respect to operational field evaluation testing, the
specification requires the agency *to determine the need for an
operational field evaluation* and to document the rationale for *no field
test.*  Specification 5100-304b S: 3.14.
[8] Because language in the specification regarding changed or modified
products provided the agency with the discretion to waiving burn and
visibility testing in this case, we need not address the parties*
arguments addressing the question of whether the specification*s use of
the phrase *applicable tests* also provided the agency with the discretion
to waive burn and visibility testing.
[9] In addition, with regard to LC-95A and LC-95D, the protester
challenges the combustion-retarding effectiveness test results of LRA. 
While we find protester*s contentions in this regard to be unfounded, they
are also, however, not relevant since the agency determined that the LC-95
formulations were qualified based on the fact that they had a higher
active salt content than lot numbers 1051695-E and 1051615-E,
respectively.  The protester also raises concerns about the burn test
results of these two lot numbers, arguing that the results were unreliable
due to problems resulting from changes to the burn test chamber.  The
agency indicates that while it had made changes to its burn chamber, which
for a period yielded inconsistent burn test results, when it tested the
two lot numbers in question, the burn chamber was providing satisfactory
results.  Tr. at 246-47.  Moreover, to the extent that the protester
contends that the test results were conducted at a time when testing
conditions in the burn chamber were variable, the agency*s witness
testified at the hearing that the burn conditions in fact were more severe
and stated that *if a product burned well enough to meet the requirements
under those severe conditions, we considered them good enough . . . .* 
Tr. at 229.  The protester*s argument in this regard thus does not serve
as a valid basis for challenging the agency*s qualification of LC-95A or
LC-95D.
The protester also maintains that the agency*s reliance on the burn test
results of D75-F for the purposes of qualifying HV-R, MV-R, and LV-R was
improper because the agency failed to demonstrate that D75-F had been
successfully burn tested.  The agency*s witness, however, testified that
D75-F had been successfully burn tested and used in the field since 1984.
    Tr. at 168.  As the agency notes, the fact that D75 has been effective in
the field lends strength to the burn testing results for D75.  Thus,
contrary to the protester*s assertions, there is sufficient evidence in
the record establishing that D75 was burn tested and properly qualified. 
See also AR, Tab AI, Letter from 1984 (indicating D75 formulations met all
specification requirements and were *fully qualified*).
[10] The protester also asserts that LC-95A actually has a lower salt
content than lot number 1051695-E.  While LC-95A has a .06 percent lower
*nominal* salt content than lot number 1051695-E--a difference which the
agency indicates is not of a sufficient magnitude to affect the combustion
retarding characteristics of the products, see Tr. at 189-91--LC-95A had a
higher analyzed salt content than the salt content of the burn-tested
version of lot number 1051695-E.  According to the agency, the *nominal*
salt content is a calculated value, based on information provided by the
manufacturer of a product.  Tr. at 480-85.  The analyzed salt content is
determined from an actual laboratory measurement of the salt content of a
product.  Tr. at 481-82. 
    
[11] As support for its argument, the protester cites a study by the
Colorado School of Mines, which was conducted at the agency*s request,
examining the corrosiveness of HV-R.  This study, however, does not
contradict the relevant laboratory corrosion test results for HV-R, which
were conducted pursuant to specification 5100-304b.   
[12] The intervenor adds that storage tanks used in the field have often
been in use for extended periods of time, in some instances as long as 30
years.  See Intervenor Comments on Agency Report, Dec. 30, 2003, at 7
n.4.  In addition, the intervenor notes that the retardant stored in these
tanks changes frequently, often from one contract period to the next.  Id.
[13] The protester argues that due to the prior problems with LV-R in
terms of corrosion, the agency should have tested for corrosiveness during
its operational field evaluation of LV-R.  The protester also suggests
that the agency improperly attempted to minimize corrosion problems with
LV-R by coating or painting the inside of the storage tanks during the
course of the agency*s field test of LV-R.  As explained above, it was
reasonable for the agency to conclude that field testing of LV-R for
corrosion was unnecessary given that HV-R had been successfully field
tested for corrosion.  Because testing for corrosion was not part of the
LV-R field evaluation, protester*s allegations about coating of tanks,
which may have minimized corrosion, are not relevant.  Moreover,
protester*s concerns about the coating of tanks are unfounded since the
agency testified that it is not an unusual practice. 
Tr. at 417.  According to the agency, many of the tanks used by bases are
recycled and it is for the base and the contractor to decide whether to
coat the storage tanks.  Id. 
[14] The protester also raises concerns about the recirculation and mixing
of LV-R and MV-R, highlighting the recirculation problems with these
products during field testing in 1991 and the fact that the field
evaluation test plan for LV-R indicated that recirculation and mixing of
the product would be evaluated.  The agency concedes that recirculation of
these products has been *an intermittent problem.*  AR, Tab I, supra. 
However, because recirculation is subject to many factors, the agency
determined that it would conduct *targeted, extensive* lot assurance and
quality assurance (LAQA) testing during the first season of field use of
these qualified products.  Id.  It should be noted that recirculation or
mixing of the product is not a qualification requirement per se and that
these products all passed the specification*s storability testing
requirements.  Consequently, there is nothing to suggest that the agency
acted unreasonably in deciding to qualify these products and address the
recirculation issues through the LAQA process.
[15] For example, because the laboratory corrosion tests for the products
were *very good,* the agency concluded that additional field corrosion
tests were unnecessary.  Tr. at 193. 
[16] The FAR does not define the terms *public* or *commercial open
market.*  Thus we are left to apply the ordinary meaning of these terms
and in so doing conclude that they would encompass sales to
non-governmental entities in arm*s-length commercial transactions.
[17] The protester argues that Astaris*s sales to Hunot are not sales to
the public in the commercial market because Hunot resells Astaris*s
products to the government for governmental use.  FAR S: 27.203-1,
however, pertains only to the circumstances of the sales of products, not
to how a product is ultimately used.  Thus, the fact that products sold in
commercial transactions are ultimately used by the government is
irrelevant to the question of whether the products have been sold or
offered for sale to the public in the commercial market.  This issue is
distinct from the issues concerning *commercial item* acquisitions, since
commercial items are expressly defined by whether they are for
governmental purposes or governmental use. 
FAR S: 2.101 (defining commercial item).     
[18] Moreover, it should be noted that a product*s thickener, colorant,
and corrosion inhibitor make up a small fraction of a product in its mixed
form.  The primary ingredients of a mixed product, by weight, are the
active salts (in addition to water, of course), and the primary use of the
products is fire suppression.  Thus, there is nothing to suggest that
changes in thickeners, colorants, and corrosion inhibitors, are anything
other than *relatively minor modifications* as contemplated under FAR S:
27.203-1(b)(2).