Mixed Oxidants


MIOX – Mixed Oxidants 1
Mechanisms for Lack of Swimmer’s Complaints in the Presence of a Persistent
Combined Chlorine Measurement
Wesley L. Bradford, Ph.D., Los Alamos Technical Associates, Inc., Los Alamos, NM and
Chief Scientist, Product Development for MIOX Corporation, Albuquerque, NM, and
Rick Dempsey, President, Simply Water, LLC, Houston, TX
17 June 2005
Evidence from reports in the technical literature, laboratory research, and operational experience
with the MIOX mixed-oxidant solution (MOS) in swimming pool water strongly indicates that
MOS causes steady oxidation of organic nitrogen compounds and organic chloramines, and rapid
completion of the breakpoint reaction on inorganic –N– fragments from that oxidation rather
than allowing accumulation of them (including volatile NHCl2 and NCl3) in the pool water, as is
likely the case using chlorine (to wit, the common swimmer’s complaints of “chlorinous” odors
and burning eyes when bleach/hypochlorite is used for disinfection). This steady removal of
organic nitrogen and rapid completion of the breakpoint reaction would be expected to cause the
following beneficial effects, as have been noted by swimmers in and operators of virtually all
pools using MOS as a replacement for chlorine for disinfection1:
• Maintenance of an acceptable disinfection residual at lower doses of FAC as MOS than
required using bleach;
• Nil production of chlorinous odors in air overlying the pool water and no burning eyes
among swimmers;
• Reduction or elimination of the need to “shock” the water with excessive bleach/hypochlorite
and/or persulfate to remove combined chlorine (the sum of chlorine as inorganic chloramines
and organic chloramines)2;
• Elimination of shocking, as well as a reduction in the amount of draining and replacing pool
water for management of combined chlorine concentrations causes a reduction in the rate of
accumulation of Total Dissolved Solids (TDS) in the water and a reduction in costs
associated with draining and replacing; and
• Improved clarity of the water both because the organic amine substrate would not be present
in high enough concentrations to stimulate bacterial growth and disinfection activity is
Better disinfection, although not as yet studied directly in pools, is also expected because: 1) the
MOS has been shown in numerous studies to be a better disinfectant than chlorine alone; and 2)
the bulk of the disinfection residual in the pool water is present as FAC not the combined
chlorines (chloramines as both inorganic and organic).
1 To date, more than 23 MIOX systems are in operation in the US alone treating over 44 separate pools of volumes
ranging from 1200 gallon spas to 800,000 gallon Olympic pools.
2 The experience of pool operators using MIOX MOS (replacing bleach/hypochlorite for disinfection) to date is that
the need for “shocking” has been eliminated and that the combined chlorine accumulated in a day of heavy swimmer
load is largely removed from the water overnight so long as the FAC as MOS residual is ³ 2.0 mg/L.
MIOX – Mixed Oxidants 2
Studies recently completed on two MOS-treated pools revealed three striking features:
1. Nitrate expressed as nitrogen (NO3
– -N) concentrations do NOT accumulate in the MOStreated
pool waters, in contrast to literature reports on bleach-treated pools. This observation
suggests a chemical mechanism for the degradation of organic nitrogen compounds by MOS
that is different from that of bleach;
2. The accumulated Total Organic Carbon (TOC) concentrations from the Body Fluid Analog
(BFA)i added continuously by bathers are MUCH lower than those reported in the literature
for bleach-treated pools; and
3. The cause for persistent positive biases in the Combined Chlorine (CC) measurements, which
are observed in some MOS-treated pools, is most likely slow degradation of some organic
nitrogen (organic–N) components of the BFA. This results in small concentrations being
observed (in the Total Kjeldahl Nitrogen analysis) at any time, and formation of organic
chloramines – a persistent CC measurement arising from this mechanism is equivalent to a
“nuisance residual”.
The MIOX-generated MOS has been used as a replacement for chlorine (as gas or
bleach/hypochlorite) for disinfection treatment of swimming pools since the first installation at
Longmont, CO in 1994. That early installation, while not successful for operational reasons,
resulted nevertheless in reports that the pool operators and users were very pleased with the
quality of the pool water. Complaints by swimmers of odors and burning eyes, prominent when
chlorine (form of chlorine not fully specified but most likely bleach) was used, virtually ceased
using MOS. As the numbers of pool installations worldwide has grown, similar glowing reports
on the benefits of the MOS have been received repeatedly.
Some of the benefits of using MOS for disinfection treatment (replacing bleach/sodium
hypochlorite) in swimming pool water were observed early and discussed at length by Bradfordii.
The most common swimmer’s complaint – chlorinous odors, together with itchy dry skin and
burning eyes – disappear within days of the introduction of MIOX MOS. The chlorinous odors
are due, primarily, to volatile dichloramine (NHCl2) and the burning eyes to volatile
trichloramine (NCl3), both created by the incomplete reaction of ammonia expressed as nitrogen
(NH3-N) with FAC in route to complete breakpoint which releases the nitrogen as harmless N2
gas. Commonly as well are observations in the chemistry of the pool water, lasting from several
days to several 10’s of days, consistent with a model/hypothesis involving release of biofilms
from the pool circulation system. Other improvements in the water follow, including dramatic
increases in water clarity. These observations have been made in enough pools to have allowed
development of expectations applicable to any pool starting MOS treatmentiii.
Beginning in 2002, Mr. Rick Dempsey, coauthor of this paper, the distributor for MIOX for
recreational water applications in the US, and consultant (with over 12 years experience) on
chemistry and operations to the commercial pool industry, reported these same observations
with considerable quantitative detail added, plus additional observations, in his clients’ pools
when they replaced MIOX MOS for bleach/hypochlorite for disinfection.
Chemistry and Chemical Mechanisms
The chemical mechanisms responsible for these benefits are not completely clear. Part of the
reason for this lack of clarity in knowledge stems from the fact that swimming pool maintenance
MIOX – Mixed Oxidants 3
technology appears to have reached the point that some practices can best be described as
“folklore”, that is practices for which the chemistry is either not well established, is sometimes
misunderstood by the practitioner, or is not discussed in such comprehensive texts on the
chemistry of disinfection as Faust and Alyiv or Whitev. Pool maintenance practices are generally
straightforward, typically controlling the pH to a range of 7.2 – 7.6 and a disinfectant (chlorine)
residual of no less than 2 mg/L (often an upper limit of 5 mg/L is set by a state regulatory
agency). But in many cases when discussing maintenance practices with a pool operator it is
unclear whether the disinfectant residual is as Free Available Chlorine (FAC) or as Total
Chlorine (TC)3.
The distinction between FAC and TC is important because TC includes FAC and the so-called
“combined chlorine” compounds which are chlorine reacted with an amine (nitrogen) group. The
chloramines (predominantly NH2Cl, but NHCl2, and NCl3 are often produced inadvertently) are
used widely in potable water disinfection because they provide a disinfectant residual necessary
to meet regulatory requirements and are weaker oxidants than FAC and therefore less reactive,
forming less disinfection byproducts (Trihalomethanes or THMs). They provide some biocidal
(disinfection) action but less than that of FAC. Chlorine in its various forms in potable and pool
water also impart taste and odor. White5 (p. 395-396) notes that “[T]astes and odors [T&O] from
the application of chlorine are not likely to occur from the chlorine compounds themselves up to
the [concentration] limits listed below:”
Chlorine Compound Threshold of T&O
Henry’s Law Constant1
Free Chlorine (HOCl) 20.0 mg/L 3.2 x 10-5
Monochloramine (NH2Cl) 5.0 mg/L 3.4 x 10-4
Dichloramine (NHCl2) 0.8 mg/L 1.1 x 10-3
Nitrogen Trichloride
0.02 mg/L 0.33
Also listed in the table are the Henry’s Law constants (H) for the relevant forms of inorganic
chlorine in pools under conditions typical for pools (pH ~ 7.5; Temperature ~ 28oC (~ 82oF))1.
The Henry’s Law constant (H) is the ratio of the concentration of the compound in the air above
the water to its concentration in the water at equilibrium. As the volatility of the compound from
the water rises, the numerical value of H also rises. Clearly the Threshold of T&O Complaints
decreases with increasing volatility. Also, both Free Chlorine and Monochloramine have very
low volatility, whereas Dichloramine and Nitrogen Trichloride would be considered highly
Faust and Aly4 (p. 105), however, note that NH2Cl imparts a chlorine-like odor and flavor to
potable water and NHCl2 is associated with a swimming-pool-like, bleach-like taste and odor.
Fortunately for the drinking water consumer, NH2Cl, with its higher threshold of complaint due
to taste and odor, predominates in potable water systems with NHCl2 and NCl3 forming only at
pH’s less than about 7.0, which is lower than typical, and in chlorine dose to ammonia nitrogen
3 Awareness of the distinction between FAC, TC, and combined chlorine (CC) has grown considerably in recent
years particularly within the commercial and municipal swimming pool community, and the regulatory agencies.
Many county health departments are also aware of the key role of the CC measurement in pool management and
public safety, and regulate pools on that basis.
MIOX – Mixed Oxidants 4
(Cl/NH3-N) ratios greater than about 10:1, which is much higher than the typical < 6:1 used in
potable water treatmentvi.
The combined chlorine component of TC also includes chlorine reacted with the nitrogen groups
of organic amines to form organic chloramines. The technical literature strongly indicates that
the organic chloramines have nil disinfection capability. But at least a portion of these
compounds detect as TC in standard wet chemical analyses4; thus their presence can easily
mislead a pool operator into thinking he has adequate disinfectant residual and, thus, adequate
disinfection, when in fact he does not.
While tastes and odors imparted by the organic chloramines are not discussed in recent texts,
significantly from the standpoint of pool maintenance, White5 (p. 460) notes that high Cl/NH3-N
ratios (12:1) with high organic nitrogen compound concentrations in water forms NCl3 which has
a geranium-like odor, an odor that can be taken as chlorinous. Elsewhere White5 (p. 260) states
that NCl3 “…is characterized by its pungent, chlorine-like odor”. And (p. 396) White5 notes that
NCl3 solubility in water is negligible, it will aerate with the slightest agitation, and at
concentrations [in air] too low to get a response from the olfactory system, it will cause the eyes
to tear profusely” [italics added by the authors].
Because chloramine formation is known to progress with increasing Cl/NH3-N ratios from
NH2Cl to NHCl2 to NCl3 and finally to N2 (the breakpoint), it is reasonable to expect that the
nitrogen in organic amines, after being fragmented by oxidation, releasing inorganic –N–
fragments to the water, reacts with chlorine in the same manner with progressive Cl/NH3-N dose
ratios. Thus a significant fraction of the total chloramines (responsible for the CC measured)
present in swimming pool water may occur as NHCl2 and, with increasing doses, as NCl3 in the
presence of high organic nitrogen (amine) concentrations and progressively higher Cl/NH3-N
dose ratios. Indeed, Hery and Hechtvii showed that “…roughly 90% of the swimming pool
atmosphere pollution by the chlorine species is due to nitrogen trichloride [trichloramine, NCl3]”.
[note added by the authors].
In summary, the technical literature demonstrates chemical mechanisms for formation of at least
two compounds by reaction of chlorine with organic nitrogen (amines) – NHCl2 and NCl3, – both
of which have properties that are consistent with observations of pool users of objectionable
conditions. Both have chlorinous, bleach-like odors and are volatile, slightly soluble in water,
and NCl3 causes tearing and irritation of the nose and respiratory tract at concentrations [in
air] that are below thresholds of odor. Moreover, NHCl2, NCl3 and organic chloramines detect
as TC (and CC) on commonly-used chlorine test kits but provide nil biocidal action.
Expected Chemistry in a Typical Pool Maintenance Scenario Using Chlorine (Bleach)
4 The fact that the chlorine in organic chloramines detects as CC in several analyses for FAC and TC – including the
DPD colorimetric and DPD-FAS test commonly used in testing swimming pool waters – is well established and
well known within the potable water treatment industry and the technical literature (for example see Gordon, et.al.6).
This fact appears to be much less well known in the swimming pool industry – the instructions which accompany
commonly used test kits do not note this effect although the technical staffs of the manufacturers of these test kits
will readily explain it. Considering the abundance of organic–N compounds in the BFA, and the concentrations of
organic–N compounds which would be expected to accumulate in a heavily-used pool, this lack of understanding of
the chemistry of the major test method used for pool control is stunning.
MIOX – Mixed Oxidants 5
In a typical pool maintenance scenario using bleach for disinfection, one would expect both
organic amines and, with continuing chlorination5, organic chloramines, and inorganic –N–
fragments (chlorinated as inorganic chloramines) from oxidative decomposition of organic
amines to accumulate. As the Cl/NH3-N dose ratio rises, NHCl2 and NCl3 are produced as
discussed in the background text above. The presence of these compounds would be expected to
lead at once to complaints of swimming-pool-like chlorinous odors and burning eyes, symptoms
which are well documented in the technical literature as being associated with these compounds
and typical of pools using bleach.
Reductions in water clarity with time and bather load that often occur in swimming pools using
bleach for disinfection may be due solely to the fact that biocidal activity does not increase with
increasing chlorine dose, despite accumulation of organic chloramines. Indeed, biocidal activity
may decrease because the organic chloramines provide nil biocidal activity, even though the TC
concentration would appear to the operator to increase. Moreover, the organic amines are food
substrate for bacteria. It would be expected that, as organic amines increase in concentration but
biocidal activity stays constant or decreases, certain bacteria can find a niche in the pool water,
develop into colonies, and contribute to loss of water clarity.
MIOX MOS and the Breakpoint Reaction
The MOS has been shown in laboratoryviii and subsequently in potable water systems in Texasix
and in Iowax to oxidize ammonia (NH3-N) and inorganic chloramines to nitrogen gas (N2) at
Cl/NH3-N dose ratios lower than either the theoretical (7.6:1) or the practical (8 – 10:1; in fact,
ratios as high as 15:1 have been reported to be used) ratios required to drive the breakpoint
2 NH3 + 3 Cl2 ® N2­ + 6 Cl- + 6 H+
At the Fonda, IA water treatment plant10, for example, Cl/NH3-N dose ratios as low as 5.2:1 as
MOS caused a breakpoint-like reaction leading to complete loss of NH3-N from the raw water.
The most plausible chemical explanation for this effect is that the non-chlorine components of
the MOS, rather than the chlorine, react preferentially with ammonia and chloramines causing
loss of the ammonia as nitrogen gas but reduced (compared to chlorine alone) consumption of
the chlorine.
While research on the chemical effect of the MOS on organic chloramines has not been
performed as yet, it is hypothesized and expected that the MOS, which is known to be a stronger
oxidant than chlorine alone, causes steady decomposition of the organic chloramines to inorganic
–N– fragments rather than an accumulation of them in swimming pool water6. This steady
5 It is worth noting that in larger public swimming pools, the rate of chlorine dosing is often controlled by a pH/ORP
(Oxidation-Reduction Potential) controller. The ORP sensor detects mostly the FAC component because the ORP
of chloramines and organic chloramines is much lower than that of FAC. Thus, chlorine will continue to be added
based largely on the FAC component regardless of the TC residual concentration. In home swimming pools,
chlorine tends to be added continually also through use of solid calcium hypochlorite (Ca(OCl)2) or High Test
Hypochlorite (HTH) with minimal monitoring of the disinfection residual. In either case, the accumulation of
organic chloramines scenario as discussed in text is likely to occur.
6 Another aspect of pool maintenance “folklore” that may be relevant to this discussion is that when pool water
becomes cloudy and/or complaints of odor and burning eyes become frequent, it is common practice to “shock” the
water with a large overdose of bleach. The chemical effect of the shocking would be oxidation of accumulated
inorganic chloramines to nitrogen gas by the breakpoint reaction and, probably, oxidation of the accumulated
MIOX – Mixed Oxidants 6
decomposition of organic amines to inorganic –N– fragments is followed (in the presence of
FAC as MOS) by rapid completion of the breakpoint reaction, allowing little if any accumulation
of the volatile, chlorinous-odor causing NHCl2 and NCl3 reaction intermediates. Such an effect
would be completely consistent with other information and observations accumulated over 10-
years of research and operational experience on the superior disinfection and chemical oxidation
behavior of the MOSxi.
Expected Chemistry in a Pool Maintenance Scenario Using MIOX MOS
A swimming pool maintenance scenario using MOS instead of bleach for disinfection would be
expected to show the following features due to: 1) better disinfection; and 2) steady oxidation of
organic chloramines to inorganic –N– fragments followed by rapid completion of the breakpoint
• Maintenance of an acceptable disinfection residual (both FAC and TC) at lower FAC doses
than required using chlorine;
• Nil accumulation of volatile NHCl2 and NCl3 in the water and, as a result, no chlorinous
odors in the air overlying the pool water and no burning eyes among swimmers;
• Dramatic reduction or complete elimination of the need to periodically “shock” the pool
water with excessive bleach/hypochlorite and/or persulfate to remove combined chlorine;
• Better disinfection both because the MOS is a better disinfectant than chlorine alone and the
bulk of the disinfection residual in the pool water would be present as FAC not the combined
chlorines; and
• Improved clarity of the water both because the organic amine substrate would not be present
to stimulate bacterial growth and disinfection effectiveness is increased.
Beginning in September 2004, the three pools at a Pool Complex in New Jersey using MOS for
disinfection began developing a persistent CC measurement >> 0.2 mg/L but WITHOUT the
swimmer’s complaints or other features normally associated with a high CC measurement –
obviously the commonly-accepted quantitative correlation between the CC measurement and
swimmer’s complaints was NOT present using MOS. We attempted many times between
September and November to eliminate the CC measurement by manipulating the operations of
the pool, but to no avail. It was fairly clear, however, that some constituent in the water of the
three pools was causing a persistent positive bias in the CC measurement (using the DPD-FAS
test); literature research by Bradford identified several candidate constituents. In addition,
experiments with the pools indicated that the constituent(s) causing the persistent positive bias in
the CC measurement was being ADDED with the MOS rather than being created in the pool
water, and that the constituent(s) also reacted with and was partially consumed by other
constituents of the pool water – most likely these reactions are oxidation of organic material
added in the BFA7. Samples from the three pools were collected and analyzed at a commercial
water analysis laboratory; results reported in December 2004.
organic chloramines and chlorinated inorganic –N– fragments as well. The biocidal effect of shocking is obvious.
The net result is, as would be expected, improved clarity of the water and cessation of swimmers’ complaints.
7 The CC measurement would rise to >1.2 mg/L as the MOS dose increased then decline, trending to 0.4 – 0.6 mg/L
as if any reaction stopped at a concentration of about 0.4 mg/L (as a CC measurement). The Oxidation-Reduction
MIOX – Mixed Oxidants 7
Since it was uncertain what constituent(s) could be causing the persistent positive bias in the CC
measurement, the objective of the analyses was to obtain as complete a chemical characterization
as possible, given current analytical capabilities. As it seemed fairly certain at the time [later
proved to be incorrect] that the constituent(s) of interest was an inorganic, and complete
characterization of the organic compound composition is currently impossible, the focus was on
inorganic cations and metals, and anions, plus a Total Organic Carbon (TOC) analysis to assess
the TOC status of the pool waters relative to recent published findings.
• Cations/Metals – Method M201 (a version of EPA Drinking Water Method 200.8); ICP/MS
broad scan metals;
• Anions – EPA Methods 300.0, 300.1 (BrO3
-): Ion Chromatography; and L500 (ClO4
Liquid Chromatography with quadrupole Mass Spectrometry (LC/MS/MS); and
• Total Organic Carbon – Standard Methods 5310C, UV persulfate digestion.
In addition, the director of R&D and the inorganic department manager at the analyzing
laboratory examined the anion chromatograms for unquantified peaks [commercial analytical
laboratories quantify only those peaks and constituents for which analytical standards have been
established]. Significant unquantified peaks would have been investigated further by more
specialized methods; however, although other peaks were noted, they were not quantitatively
significant and a tentative identification was made, so no further investigation was conducted.
The broad-scan metals analysis showed no unusual concentrations of metals and was dropped in
later studies. Results of other analyses are discussed below.
A few months later, a similar occurrence of a persistent CC measurement (0.2 – 0.4 mg/L) but
again WITHOUT the common swimmer’s complaints appeared in a Swim School pool in North
Carolina also using MOS for disinfection. A sample was collected in April 2005 and analyzed at
the same laboratory for the same constituents (sans the broad-scan metals), plus a special-request
Total Kjeldahl Nitrogen (the sum of ammonia (NH3-N) and organic-N) which had been thought
to be unimportant in the analyses of samples from the Pool Complex and had not been requested.
Results of the analyses of samples from the two facilities are listed below.
FACILITY Pool Complex
Pool Lap Therapy Spa
Analyte Units 166
20 kgal 1200
190 kgal
Sodium (Na+) mg/L 130 230 300 370
Potassium (K+) mg/L na na na 9.5
Lithium (Li+) mg/L 4.5 2.3 4.7
Rubidium (Rb+) mg/L 0.021 0.018 0.022
Calcium (Ca2+) mg/L 98 63 57 88
Magnesium (Mg2+) mg/L 1.7 1.2 3.1 3.0
Barium (Ba2+) mg/L 0.0088 0.020 0.016
Potential (ORP) reading tended to be invariant with the FAC, suggesting the presence of a separate reaction couple
which strongly influenced the ORP measurement as a mixed-couple.
MIOX – Mixed Oxidants 8
Strontium (Sr2+) mg/L 0.72 0.45 0.26
Alkalinity (as
mg/L 90
Chloride (Cl-) mg/L 760 1200 1700 620
Sulfate (SO4
2-) mg/L 75 48 44 86
Nitrate (NO3
– -N) mg/L 0.55 0.53 1.1 0.53
Phosphorus (P) mg/L 0.11 0.11 0.32
Bromide (Br-) mg/L 0.65 0.54 1.4 < 0.010
Bromate (BrO3
-) mg/L 0.14 0.21 0.22 0.032
Chlorite (ClO2
-) μg/L < 10 < 10 < 10 < 10
Chlorate (ClO3
-) mg/L 19 28 56 23
Perchlorate (ClO4
-) μg/L 4.5 16 28 13
TOC mg/L 13.6 14.8 35.2 9.32
Total Kjeldahl-N mg/L (0.058)1
1 The regulatory required Method Reporting Limit (MRL) is 0.5 mg/L and that figure was shown
on the official report. The value shown was obtained from the lab analyst; his lowest standard is
0.05 mg/L, and that was the Detection Limit.
Physical Characteristics and Bather Load Estimates for the Pool Complex
Currently estimates of bather loads are available only from the Pool Complex. Similar estimates
from the Swim School are being developed; however, based on experience with swim schools
generally, the bather load at the Swim School is expected to be similar to that of the Therapy
Pool at the Pool Complex.
Function Volume Temperature Bather
Gallons Cubic
oF bathers/m3
Lap Pool 166,000 628.3 82 0.11
Therapy Pool 20,000 75.7 88 – 90 0.44
Spa 1,200 4.5 102 – 104 2.2
The bather loads are estimated based on the following assumptions:
• The volume of fluids discharged per unit time (hour) from each swimmer is a function of
water temperature and swimmer activity – the temperature function alone is approximately 1
pint/hour at ~ 80oF, 2 pints/hour at ~ 90oF and 3 pints/hour at ~100oF and normal activity for
that temperature, i.e. moderate lap swimming in the Lap Pool and swimming lessons of
moderate activity in the Therapy Pool. These approximate volumes of body fluids are used as
“equivalent bather” weighting factors as described below to normalize to published findings
of pool chemistry as a function of bather loads.
• Swimming teams using the Lap Pool are exceptionally active, however – the “equivalent
bather” is given a weighting factor of 2 consistent with the level of activity.
• Pool operation for 15 hour/day (0600 – 2100 hours), 7 days per week as a normal operation –
all pools are closed from time to time for maintenance, of course.
MIOX – Mixed Oxidants 9
Bather loads are estimated as follows: for perspective, a bather load of 0.5 bathers/m3 hr is
considered exceptionally heavy.
hours bath
Lap AM 100 1.5 1 150
PM 150 0.75 1 112
PM team 150 2.5 2 750 1012 0.11
Therapy Day 250 1.0 2 500 500 0.44
Spa Day 100 0.5 3 150 150 2.2
The chemistry of each pool is maintained within narrow limits of hardness and alkalinity by
manual additions of chemicals when necessary. The pH is maintained in the range 7.4 – 7.6 by
continuous monitoring and addition of acid (HCl) by pH controller. The FAC is maintained at
about 2.0 mg/L ± by continuous ORP monitoring and addition of MOS (previously bleach). The
controller can be programmed to increase the ORP set point at specific times to add more MOS
(aka super MOS dosing) increasing the FAC; this operation is conducted often at night during off
hours, allowing the excess FAC to be consumed by oxidation reactions and reach a concentration
allowed for operation (typically  4 mg/L) by morning opening. FAC and CC measurements are
made manually using the DPD-FAS drop-wise titration method several times during an operating
day and the measurements used to adjust the ORP set point for MOS addition; the method has a
sensitivity of 0.2 mg/L8.
Discussion of Analytical Results
Alkali Metal Cations (Na+, K+, Li+, Rb+)
Sodium (Na+) and potassium (K+) normally constitute the bulk of the alkali metal cations in a
water sample. K+ is not detected by ICP so is not reported for the Pool Complex; the
concentration seen in the Swim School sample is typical even of potable waters. The high
concentrations of Lithium (Li+) reflect the use of lithium hypochlorite (LiOCl) to supplement
MOS at the Pool Complex, and Rubidium (Rb+) is most likely a contaminant of both bleach (also
used occasionally when MOS ran short) and LiOCl. Lithium concentrations are highest in the
Lap Pool and the Spa, reflecting the tendency for LiOCl to be used more in those pools.
Alkaline Earth Metal Cations (Ca2+, Mg2+, Ba2+, Sr2+)
8 The measured ORP is known to be influenced by the major ion composition of the water (the effect of Cl- is
predicted theoretically, for example, as a 30 mV decrease per decade increase in Cl- concentration and the ORP is
observed to decrease as the Total Dissolved Solids (TDS) concentration increases) and dramatically influenced by
the presence of both inorganic and organic chloramines; the ORP at a given measured FAC drops by several 10’s of
millivolts when cyanuric acid is added to the water to a concentration of a few 10’s mg/L. During the startup period
using MOS in fresh pool water, the ORP-FAC relationship initially approached that predicted theoretically in
contrast to that observed using bleach; however, with continued MOS use without changing the pool water, the ORP
measurement tended to be independent of the FAC, suggesting the increasing presence of a separate oxidationreduction
couple which tended to dominate the ORP, an effect on ORP from increasing TDS, or other factors not yet
MIOX – Mixed Oxidants 10
These four divalent cations are all detected as hardness. Calcium (Ca2+) and magnesium (Mg2+)
are typically the major components of hardness in waters; the smaller concentrations of barium
(Ba2+) and strontium (Sr2+) probably arise as contaminants in the chemicals used for hardness
Major Common and Minor Anions
Chloride (Cl-) is expected to be “conservative” in waters, i.e. in the absence of volatile losses of
chlorine as Cl2 or as the inorganic chloramines, it does not engage in reactions leading to net
losses (the same may be said for chlorate (ClO3
-) and perchlorate (ClO4
-) discussed later).
Because the major source of Cl- in the pools is the NaCl in MOS, as well as in the other chlorinebased
disinfectants used, the concentrations are expected to be proportional to the total MOS and
other disinfectant doses since last draining. Any quantification of the proportionality to MOS
dose exclusively, however, is confounded by the use of lithium hypochlorite (LiOCl)
preferentially in the Lap Pool.
Sulfate (SO4
2-) is normally expected to be “conservative” in waters as well except that it can
engage in precipitation reactions, especially with Ca2+ to form gypsum (CaSO4), leading to a net
loss of both Ca2+ and SO4
2-. Gypsum (CaSO4) is one of the few compounds in nature showing
reduced solubility with increase in temperature (a feature which causes gypsum scale formation
in boilers). Thus the differences (decreases) in both Ca2+ and SO4
2- from the coolest to warmest
of the three pools of the Pool Complex may result from more precipitation of gypsum in the
warmer pools, with the gypsum being removed in the filters.
Nitrate (as nitrogen, NO3
– -N) is a curiosity the evaluation of which strongly suggests that there
are different reaction pathways between the reactions of FAC as MOS and FAC as bleach (or
LiOCl) with organic nitrogen compounds and ammonia. Judd and Bullockxii reported in their
experiments with a model pool dosed with BFA at varying rates and holding the FAC (as bleach)
at a concentration at 2.0 ± mg/L that NO3
–N accumulated in the pool water to very high
concentrations (15 mg/L was the highest observed) and accounted for 4 – 28% of the dosed
amino nitrogen (both NH3-N and organic-N). Not unexpectedly, the rate of NO3
–N accumulation
was found to increase with bather load as shown in the figure below (from Judd and Bullock
The fact that the NO3
–N concentrations in the four pools were 1) very low compared to the
concentrations accumulated in the Judd and Bullock12 model pool despite extremely high bather
loads, and 2) virtually invariant among the pools indicates that, using MOS, the load of organic-
N from the BFA is ultimately discharged as nitrogen gas (N2) through the breakpoint reaction of
inorganic –N– fragments and NH3-N.
Nevertheless, the sharp contrast between the findings of these analyses and those of Judd and
Bullock is clear indication of a different reaction pathway in waters treated with MOS compared
to those treated with bleach. This, in turn, indicates the presence of at least one non-chlorine
oxidant or catalyst in MOS not present in bleach.
MIOX – Mixed Oxidants 11
Phosphorus (P) — the sources and chemical behavior of phosphorus are not known well enough
at this time for any comment to be useful. However, the chemical form is probably orthophosphate
3-) as H2PO4
-; the lead chemist at the analyzing laboratory noted an indication
of phosphate on the anion chromatograms but the concentrations were below the level of
quantification (~ 0.2 mg/L as P).
Bromide (Br-) and Bromate (BrO3
-) — the presence of these constituents is expected; bromide is
known to be present in the salt used for MOS generation and Gordonxiii had noted the presence of
small concentrations of bromate in the MOS generated by a small MIOX system, concluding that
it was being produced by electrolysis in the MIOX cell from bromide in the salt. Occasional
analyses of MOS treated potable waters containing Br- have shown that MOS does not make
– in the treated water at least to detectable concentrations and probably not at all.
Weinberg et.al.xiv and more recently Brown, et.al.xv reported bromate in bulk bleaches used for
potable water treatment. Brown et.al. report concentrations of 0.244 – 0.310 μg BrO3
– /mg FAC
(MCL  10 μg/L). At these concentrations in bleach and given typical FAC doses for potable
water treatment (~ 5 mg/L), BrO3
– concentrations due to bleach disinfection are not likely to
approach the MCL. However, in swimming pools which receive continual doses of FAC as MOS
or as bleach, both Br- and BrO3
– concentrations would be expected to accumulate. The authors
have inquired with the National Spa and Pool Institute/Association of Pool and Spa Professionals
(NSPI/APSP) for guidelines or enforceable regulations on maximum BrO3
– concentrations in
pools; none have been issued, although the NSPI/APSP Recreational Water Quality Committee
has recently discussed the accumulation of other constituents associated with disinfectants. For
the moment at least, there appears to be no concern for BrO3
– or other constituents in pools or
other recreational waters.
Moreover, it is clear from the analysis of the Swim School sample that accumulation of Br- and
– can be virtually eliminated by using a low Br- salt for MOS generation; the Br-
MIOX – Mixed Oxidants 12
concentration in the salt used at the Pool Complex was estimated at > 270 ppm (mg/kg) whereas
that used at the Swim School was known to be < 100 ppm.
High concentrations of Br- lead to dominance of Br2/HOBr/OBr- as the disinfectant – so called,
“bromine pools”. Inadvertent addition of Br- into a bleach-treated pool under ORP control is also
known to affect the ability of the ORP controller to prevent development of conditions typical of
a “bromine pool”. Bromine pools are not known to develop chorinous (or brominous) odors
because the bromamines NHBr2 and NBr3, analogous to the chloramines, have low volatility.
However, the coauthor’s (Dempsey) experience is that bromine pools tend to require more
chlorine because the Br2/HOBr/OBr- forms are not as strong oxidants; the water takes on a
cloudy, gray, dull-green appearance and swimmers complain of itchy skin and a condition
(probably misnamed) known as “bromine rash”, all thought to be due to accumulation of
bromamines (but possibly also due to accumulation of organic material and organic
bromamines). These conditions require a chlorine or non-chlorine shock to remove. Operators
tend to use more chlorine (as bleach) in bromine pools in order to maintain other features of the
pool water, most notably water clarity. Dominance of disinfection by Br2/HOBr/OBr- is known to
occur at Br- concentrations around 5 mg/L; there is little experience with the effects of Br- at submg/
L concentrations on pool management.
It is reasonable to ask why, since BrO3
– (and perforce Br-) are also present in bleach, chemical
behavior typical of a “bromine pool” is not regularly seen in bleach treated pools! The answer
may be that bleach treated pools are typically drained much more frequently for control of CC
and TDS than are MIOX MOS treated pools; thus the BrO3
– and Br- concentrations cannot build
to the levels seen in these analyses. The experience to date in pools on which the coauthor
(Dempsey) has consulted is that bleach treated pools are drained 4 to 6 times for each drain of a
MIOX MOS treated pool because the driver to control CC is not present in the MIOX MOS
treated pools. Instead they tend to be drained due to TDS buildup and for structural and piping
maintenance. And the rate of TDS accumulation in MOS treated pools is also lower than in
bleach treated pools because there is no need for shocking chemicals (excess bleach or
persulfate) in MOS treated pools.
Chlorite (ClO2
-), Chlorate (ClO3
-), and Perchlorate (ClO4
-) — The low (< 10 μg/L) chlorite
-) concentrations provide further verification that MOS does not contain chlorine dioxide
The chlorate (ClO3
-) and perchlorate (ClO4
-) concentrations in the three pools of the Pool
Complex increase in the same direction as the bather loads and the MOS dose. This is expected
because MOS is known to contain small concentrations of both chlorate (~ 13 – 33 μg/mg FAC)
and perchlorate (~ 0.2 μg/mg FAC). However, the concentration ratios between the pools are not
the same for the two constituents – the chlorate concentration appears to be high relative to the
perchlorate concentration in the Lap Pool. This apparent aberration is likely due to the use of
LiOCl referentially in the Lap Pool; it is likely that LiOCl contains chlorate as a contaminant.
The chlorate and perchlorate concentrations in the Swim School sample are like those of the
Therapy Pool of the Pool Complex, supporting the expectation (to be confirmed) that the bather
load at the Swim School is similar to that of the Therapy Pool.
The last survey (known to the authors) of bleaches used for potable water disinfection showed
chlorate concentrations ranging from 0.018 – 2.6 mg ClO3
-/mg FAC, with a median of 0.11
mg/mg FACxvi; even the median concentration found in this survey is much greater than the
MIOX – Mixed Oxidants 13
concentration range determined in MOS (0.013 – 0.033 mg/mg FAC). Despite the occurrence of
chlorate in bleach, probably due to the absence of an EPA MCL for chlorate in potable waters,
no concerns for the accumulation of chlorate in pools has been expressed by NSPI/APSP.
Moreover, despite recent attention being devoted to perchlorate contamination of potable water
supplies and the high likelihood that it is present in commercial bleaches as well, no MCL has
been proposed, and no concern has been expressed by NSPI/APSP for the accumulation of
perchlorate in pools.
Organic Compounds
Total Organic Carbon — Judd and Bullock12 found that the Total Organic Carbon (TOC)
concentration reached a maximum, or “equilibrium”, value as a function of the bather load in the
pool, with the FAC concentration being held constant at 2.0 mg/L ± as shown in the figure below
(from Judd and Bullock).
This finding is counterintuitive; one would expect continued accumulation of TOC as a function
of time at any given bather load. But the fact of a maximum concentration as a function of bather
load suggests that the reaction between FAC and organic material at fixed FAC is a function of
the organic concentration, i.e. as the organic concentration increases, the rate of reaction adjusts
to balance the organic loading rate through the BFA. Mass action requires that the dose of FAC
as bleach would have to increase as well with increasing BFA and increasing reaction rates;
unfortunately, Judd and Bullock do not quantify the FAC as bleach doses. Moreover, the reaction
between FAC and organic material leads to complete mineralization of the organic to CO2 and
H2O; any fragments of organic material would have been detected as TOC.
The bather loads for the three Pool Complex pools applied to this figure would lead to the
conclusion that the TOC observed in the Lap Pool (13.6 mg/L at a bather load of 0.11 bathers/m3
hour) is very close to that predicted in the figure, but that the TOC concentrations observed in the
Therapy Pool (14.8 mg/L at a bather load of 0.44 bathers/m3 hour) and the Spa (35.2 mg/L at a
MIOX – Mixed Oxidants 14
bather load of 2.2 bathers/m3 hour) are much lower than predicted in the figure; indeed the latter
two bather load values are off the chart. Since the Therapy Pool and the Spa were treated almost
exclusively with MOS, whereas the Lap Pool received LiOCl preferentially when MOS was in
short supply, these results suggest that FAC as MOS is considerably more effective than FAC in
any other form at decomposing organic material from the BFA. In addition, the Therapy Pool
was subjected several times to overdosing with MOS overnight, and each time the excess FAC
was consumed before the pool opened at 0530 the following morning (in some experiments,
consumption of the excess FAC was complete within one hour). In the coauthor’s (Dempsey)
experience, this NEVER occurs using bleach for overdosing; the consumption of FAC as bleach
requires much longer and the FAC remaining in the morning after overdosing with bleach the
previous evening often exceeds the upper limit for opening and must be quenched with
The TOC concentration in the Swim School sample was lower than that in any of the samples
from the Pool Complex at a bather load expected to be like that of the Therapy Pool of the Pool
Complex. However, the Swim School pool is not known to have been overdosed with MOS.
Therefore, the MOS overdosing conducted in the Therapy Pool at the Pool Complex is likely to
have been irrelevant to the TOC concentration observed there; i.e. the same total MOS dose
would likely have been used with either regular maintenance of the FAC or with occasional
overdosing at night. The conclusion is unencumbered — MOS maintained a MUCH lower TOC
than did bleach (in the Judd and Bullock12 model pool).
Total Kjeldahl Nitrogen (TKN) — unfortunately, no literature reports are available for
comparison with the results discussed here. However, the TKN value of 0.058 mg/L in the Swim
School sample seems remarkably low considering the predominance of organic amines (i.e.
organic-N) in the BFA, and lower (by a factor of 28) than would be expected from the TOC
concentration9. This finding strongly supports the conceptual model that MOS rapidly and
preferentially attacks the organic-N compounds, reducing them to inorganic –N– fragments [the
absence of swimmers’ complaints in MOS-treated pools strongly suggests rapid discharge of the
inorganic –N– fragments as N2 through the breakpoint reaction as discussed earlier].
Nevertheless, the presence of TKN at 0.058 mg/L is sufficient to explain the presence of a
persistent CC measurement. The –N– moiety in the organic-N compounds of the BFA would be
expected to be fully chlorinated with one chlorine atom for each –N– moiety at a Cl/N mass ratio
of 5:1, and additional chlorines (up to 2) may attach to each –N– moiety in the presence of
abundant FAC in the water (as is the case for properly operated swimming pools). Therefore, a
measured CC concentration of 0.3 – 0.6 mg/L (0.058 x 5 or 10) would be expected to have been
observed in this water sample; in fact, the measured CC concentration at the time the sample was
collected was ~ 0.4 mg/L.
A persistent CC measurement in the absence of chlorinous odors from volatile dichloramine
(NHCl2) and trichloramine (NCl3) is commonly considered a “nuisance residual”. This
9 The expected molar abundance of carbon, nitrogen, and phosphorus in organic material is the well-known Redfield
Ratio – C106N16P (the molar abundance of –N– to –C– is actually higher in the organic material of the BFA, but use
of the Redfield Ratio is sufficient to make the argument). Therefore, the mass of organic-N per unit mass of Total
Organic Carbon (TOC) would be expected to be approximately this molar ratio multiplied by the ratio of the
molecular weights of N and C, i.e. (16/106)(14/12), or ~ 0.176 N/C. Using this evaluation, the TKN concentration
would be expected to have been 1.64 mg/L (0.176 x 9.32 mg/L (TOC)).
MIOX – Mixed Oxidants 15
evaluation strongly suggests that nuisance residuals are due to organic-N compounds found in
the BFA which are slower than others to fragment.
Geographical (and possibly temporal) variations observed in the occurrence of a persistent CC
measurement in the absence of chlorinous odors (the nuisance residual) in MOS-treated pools
cannot be explained at this time. However, it is likely that they are related to the composition of
the BFA and, given geographical and seasonal variations in diet, geographical and temporal
variations in the composition of the BFA, should they be shown to occur, would not be
i Judd, S.J. and S.H. Black, 2000, “Disinfection By-Product Formation in Swimming Pool
Waters: A Simple Mass Balance”, Wat. Res., 34(5):1611-1619.
ii Bradford, W.L., 2004, “Some Benefits of the Use of Mixed Oxidants in Swimming Pools:
Speculation on Chemical Mechanisms for Lack of Swimmer’s Complaints Using Mixed-Oxidant
Solution”, Los Alamos Technical Associates, Inc., Los Alamos, NM, 20 January 2000, last
updated 02 January 2004.
iii Bradford, W.L and R. Dempsey, 2004, “General Observations and Expectations on Startup of
MIOX”, Los Alamos Technical Associates, Inc., Los Alamos, NM, and Simply Water, LLC,
Houston, TX, 29 April 2004.
iv Faust, S.D. and O.M. Aly, 1998, Chemistry of Water Treatment, Second Edition, Ann Arbor
Press, Chelsea, MI, 581 pp.
v White, G.C., 1992, The Handbook of Chlorination and Alternative Disinfectants, Third Edition,
Van Nostrand Reinhold, New York, NY, 1308 pp.
vi Gordon, G., W.J. Cooper, R.G. Rice, and G.E. Pacey, 1992, Disinfection Residual
Measurement Methods, Second Edition, AWWA Research Foundation and American Water
Works Association, Denver, CO, 889 pp.
vii Hery, M. and G. Hecht, 1998, “Occupational Exposure to Chloramines in Swimming Pools
and Vegetable Processing Plants”, presented at the 2nd International Conference on Pool Water
Quality and Treatment, March 4, 1998, Cranfield University, Bedfordshire, UK.
viii Bradford, W.L. and R.I. Cisneros, 1995, “Oxidation of Ammonia in Aqueous Ammonia
Solutions Using the MIOX Mixed-Oxidant Solution at Chlorine Concentrations Near
Stoichiometric Equivalency with Ammonia Nitrogen”, Los Alamos Technical Associates, Inc.,
Los Alamos, NM, LATA/MX-95/0015
ix Daniel, E., 1995, “Pilot Study/Engineering Report for ‘MIOX On-Sight [sic] Mixed Oxidant
Generator for Cash Water Supply Corporation’”, Cash Water Supply Corporation, Greenville,
TX, July, 1995.
MIOX – Mixed Oxidants 16
x Kuehl, N.R., 2000, “Mixed Oxidant Ammonia Removal Pilot Project”, presented at the 25th
Annual Conference of the Iowa Rural Water Association, Kuehl and Payer, Ltd., Storm Lake,
xi Bradford, W.L., 1998, “The Differences Between On-Site Generated Mixed Oxidants and Sodium Hypochlorite”,
Los Alamos Technical Associates, Los Alamos, NM and MIOX Corporation, Albuquerque, NM, last updated 9
January 2004.
xii Judd, S.J. and G. Bullock, 2003, “The Fate of Chlorine and Organic Materials in Swimming
Pools”, Chemosphere, 51:869-879. [also presented at the Aquatic Health Conference, 2-4
October 2004, Atlanta, GA]
xiii Gordon, G.L., 1998, “Electrochemical Mixed Oxidant Treatment: Chemical Detail of
Electrolyzed Salt Brine Technology”, prepared for the U.S. Environmental Protection Agency,
National Risk Management Laboratory, Cincinnati, OH, May 1998.
xiv Weinberg, H.S., C.A. DelComyn, and V. Unnam, 2003, “Bromate in Chlorinated Drinking
Waters: Occurrence and Implications for Future Regulations”, Environ. Sci. Technol.,
xv Brown, R.A., D.A. Cornwell, and M.J. MacPhee, 2004, “Trace Contaminants in Water
Treatment Chemicals: Sources and Fates”, J. Am. Wat. Wks. Assoc., 96(12):111-125.
xvi Bolyard, M. and P.S. Fair, 1992, “Occurrence of Chlorate in Hypochlorite Solutions used for
Drinking Water Disinfection”, Environ. Sci. Technol., 26(8): 1663-1665


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: