Anotec Technical Paper - Combined Sensory And Analytical Methods

2009 reprint

 

Combined sensory and  analytical methods to formulate,  develop and determine viable  odour control technologies.

Victoria Zavras  Anotec Pty Limited.                             Presented in Singapore at the IWA Odour  Conference   

Combined sensory and analytical methods to formulate, develop and determine viable odour control technologies. V. Zavras*

* Anotec Pty Ltd, Odours Research Unit, 30-32 Chegwyn Street, Botany, New South Wales, 2019, Australia (E-mail: [email protected])

Abstract Odour complaints transpire with enough regularity and enough force to warrant urgent action by regulatory authorities and industries concerned. This type of pressure inadvertently leads the operator to investing in quick-fix odour control remedies which more often than not fail to perform as promised by the supplier. This paper shows the intrinsic worth of olfactometry and analytical measurements as a combined approach in assessing the efficacy and viability of odour abatement technologies/chemicals with the view of deterring facility operators from adopting the “band-aid” approach. Used in this study were generic and site-specific formulations as developed by Anotec. As an added interest, Anotec formulated a masking agent and included it in the overall results. The results clearly negate the concept of a “broad spectrum” odour control solution. The combination of olfactometry and analytical methods effectively present the fate of an odorous emission when challenged with various odour abatement technologies assisting in the decision making stages of odour control implementation at any give site or industry. Keywords Abatement; analytical; Anotec; combined; odour control; olfactometry

INTRODUCTION Exhaust emissions usually contain significant amounts of various odorous components. The perceived odour is due to the effect of all compounds in the mixture and depends upon the concentration and odorous qualities of the component compounds. Therefore, it is logical that an odorous gas sample be characterised in two ways. Firstly, in the analytical expression, whereby the components are identified and their concentrations are measured and secondly, in the sensory expression, where the human response to the odorous mix is measured. Twelve years ago the company observed that there were no concrete independent strategies for assessing the impact of malodours on the residential community. The advantages of such strategies would have enabled them to predict their potential odour impact and for testing odour abatement technologies prior to commitment and implementation. This paper explains why the combined measurement approach was chosen and summarises the results obtained from research and commercialised treatment conducted between 1991 and 2003. At the same time, Anotec formulated a masking agent as a point of interest and for comparative reasons. In turn, the said research was dual in its aim as it examined the efficacy and viability of various odour control treatments formulated by Anotec as well as putting to the test efficacy of any masking agents. BACKGROUND In the early 1990’s odour control products were very much in demand. This saw the influx of numerous chemical products professing to be “true” odour neutralisers and not masking agents. These products claimed to eliminate odours at the source via atomising equipment and were deemed very attractive by regulators and operators alike due to the ease of application and low costs involved. Based on these vague and grandiose claims, the facility operator would (mostly out of desperation due to pressure from regulators and the community) procure the services of the odour

neutralising firm, install the system and test its efficacy based on trial and error application. This method was time consuming and expensive considering the number of times the supplier of the odour neutraliser had to re-visit the plant as odour complaints were still being received. In the end the odour neutraliser was taken out and the operator would have to start all over again – more often than not with another odour neutraliser based on its sales pitch and claims of odour elimination. The question posed by Anotec was “do these products really work?” and “How can we assess this scientifically and logically?” The term “Odour Neutraliser” was widely used at this time and was even specified to prospective bidders in odour control tenders, The City of Los Angeles, (1992, 1998). The said bid was entered into by Anotec with it’s formulated “odour neutraliser” along with thirty other companies for testing via Odour Science & Engineering Inc. laboratory. Anotec won the tender with 99.9% removal efficiency, Odour Science & Engineering (1992). Having said this, the question remained whether Anotec would also work in other industries. There was also the issue of testing protocol and the dilemma of best presenting data showing the efficacy of an odour control chemical. The following case studies include the initial design of a laboratory simulated wind tunnel experiments, Jiang, Stone and Zavras (1993), and on site sample collection from various odour sources. All case studies tested used abatement products using the various Anotec formulations as well as formulated odour neutraliser/masking agents as per supplier’s data sheet and dilution instructions. The application equipment varied between air atomising and high pressure hydraulic nozzles.

METHODOLOGY FOR TREATING ODOUR SOURCES The use of chemical fingerprinting has a number of stages to go through in order to satisfy quality control criteria and enable confident use of chemical/odour removal efficiencies and the implications the data has for various treatment options (Stone 1994). This methodology demonstrates that a reasonable correlation exists between organoleptic response and the organochemical response. This means that the odour strength can be described in chemical concentration terms Kaye and Stone (1994).

METHODS GC/MS Instrumental chemical analysis can provide quantitative information on the components in the gaseous mixture. This information is necessary in order to assist in the identification of the odour source, to determine whether chemical specific emission limits are compliant with any set regulations, preparing an odour profile database for a specific site during various activities to assist in the identification of the main odour contributing processes for improvement and, finally, to establish best available technology for odour mitigation applications. Dynamic Olfactometry The human response to a malodour cannot be readily determined or calculated solely based on a chemical fingerprint. Olfactometry reveals the human response to a particular odour. This measurement approach is important when addressing odour annoyance issues. Testing laboratories and Standards Anotec commissioned external independent laboratories for all odour measurements relating to the case studies mentioned in this paper. Peak time for sewage odour emissions was conducted by Sydney Water’s trading arm Australian Water Technologies’ EnSight division (Boston, 1994).

GC/MS. In all instances, the technique of thermal desorption GC-MS was employed using the most advanced equipment currently available in Australia. Determination and identification of chemical odour components and concentrations, chemical fingerprinting was performed solely by Dr. D.J.M. Stone of ANSTO’s Environmental Science Program, Lucas Heights. Dynamic Olfactometry. Olfactometry data cited in this paper was conducted using either the Dutch Standard NVN 2820 or the current AS/NZ Standard 4323.3-2001. Laboratories used include but are not limited to The University of New South Wales, Centre for Wastewater Treatment’s Odour Laboratory and CH2MHill.

CASE STUDIES Laboratory – Simulated wind tunnel experiment – sewage odours (Part 1) A laboratory experiment was conducted in the first instance due to the lack of testing protocol for the assessment of odour abatement chemicals in Australia between 1991 and 1994. A simulated stack and odorous air generator was set up at the ANSTO Laboratories at Lucas Heights. Air samples were subsequently transported to the Centre for Wastewater Treatment’s Odour Research Laboratory at the University of New South Wales. The synthetic odorous mixture was introduced at a rate of 20 Litres per minute. The odorous air mixture was generated using a permeation device and the composition was of acetaldehyde, butanol, ethylamine and hydrogen sulphide. The odorous air was introduced continuously. The control, using plain water spray was tested first, then the odourless Anotec formulation followed by a formulated fragrant Anotec product. All samples were collected from the open end of the simulated stack. The odorous air (raw odour) was sampled initially prior to the operation of the spraying equipment. The spray was operated for five minutes during each subsequent test and the three following samples were collected during operation of the spray. There were intervals of about thirty minutes between successive tests. The dynamic olfactometry was undertaken generally in accordance with the Dutch standard (in the absence of an Australian Standard at the time). A panel of eight observers were used to test each of the samples. Calculation of results was done automatically by the Dravnieks method. Analysis with GC/MS was performed concurrently with the collection of each sample as prepared by olfactometry testing. This ensured that the results obtained from the chemical analysis would be directly correlated with the olfactometry results. Table 1. Olfactometry Results – Synthetic Sewage Odours Testing Parameter Odour Strength (Odour Dilution Units) Odorous Air Mixture 18,227 Raw Odour + Plain Water 4,092 Raw Odour + Odourless 1,736 Raw Odour + Fragrant 6,190

Removal Efficiency (%) n/a 78 90 66

These results showed a marked reduction with water and even more so with the odourless formulation. However the fragranced product was shown to be less effective. It was expected that the fragrance would contribute to the perceived odour and a proper assessment of this product could not be achieved. It was anticipated that the chemical analysis would enable removal efficiency for each of the compounds while choosing to ignore the fragrant components of the product. This selective ability cannot be achieved by sensory methods, Zavras (1994).

Table 2. GC/MS Results – Synthetic Sewage odours Component (ppm) Raw Odour Control – Water Hydrogen Sulphide Acetaldehyde Methylamine 2-butenal Butanol

0.566 9.818 3.000 5.630 40.540

0.068 0.037 0.500 2.770 14.590

Anotec – Odourless 0.000 0.000 <0.05 0.770 0.790

Odour Hydrogen Sulphide Acetaldehyde Methylamine 2-butenal Butanol Total Odour

2832.35 1636.36 300.00 938.29 506.76 6214

341.47 6.11 50.00 461.71 182.33 1042

0 0 0 128.37 9.84 139

Anotec – Fragranced 0.000 0.000 <0.05 0.260 0.320

0 0 0 43.91 4.02 48

Odours resulting after spraying with either the odourless or fragranced solutions showed a total elimination of hydrogen sulphide, acetaldehyde and methylamine compared to the raw odour. However, both the Butanol and 2-butenal were still present at considerably reduced concentrations after treatment with the spraying system. In this instance, the fragranced component in the formulation was ignored in the calculation of results since the analysis was required to be confined to the removal of the odorous components generated from the permeation system. Laboratory – Simulated wind tunnel – Sulphur Dioxide (Part 2) In this test technical grade sulphur dioxide was prepared at ANSTO. In this case each sample was analysed immediately after collection and were then tested again after 24 hours. The odour control products used were based on Anotec’s formulation for sulphur dioxide alone and the fragrant component was formulated using Zwaardermaker’s principle of odour pairing. Table 3. Olfactometry Results – Sulphur Dioxide

Testing Parameter

Odour Strength (ODU) Initial 87,690 13,153 4,384 6,028 11,250

Odour Strength (ODU) 24hr

Removal Efficiency (%) Initial

Removal Efficiency (%) 24hr

Raw Odour (SO2) 65,767 SO2 + Water 17,538 85 73 SO2 + Odourless 4,384 95 95 SO2 + Fragranced 4035 93 95 SO2 + Masking 18,540 87 78 agent These results show again that water contributes to the overall odour reduction of sulphur dioxide. However, over a period of 24 hours its effect “wore off” as the odour dilution unit increased when tested again. The odourless formulation remained static and did not change over a period of 24 hours. The fragranced formulation worked better than water but less effective than the odourless. When tested 24 hours later the removal efficiency was the same as the odourless formulation. It was surmised that the volatile components in the formulation biodegraded overnight thus not influencing the olfactometry reading. The masking agent showed a similar reduction in odour as per the water sample. When tested again after 24 hours we saw that the result was worse than the water

alone prompting the thought that the masking agent was dissociating from the original odour. The original odour was decreasing in odour strength due to its natural degradation process. Table 4. GC/MS results – Sulphur Dioxide Removal Efficiency (%) Initial

Component (ppm)

Removal Efficiency (%) 24 hrs

Initial result 24 Hours later SO2 9.93 7.15 SO2 + Water 1.05 1.19 89 88 SO2 + Odourless 0.49 0.49 95 95 SO2 + Fragranced 0.74 0.49 93 95 SO2 + Masking agent 1.68 2.33 83 76 Of interest here is that the Anotec formulation was developed based on sulphur dioxide alone. The fragranced component seems to have biodegraded over a period of 24 hours leaving behind the odourless formulation. Although the masking agent showed some reduction, it did not perform as well as the water. This means that using water alone will achieve the same result if using a masking agent. These results show a direct correlation between the chemistry and the olfactometry as they both show similar removal efficiencies and pattern of each testing parameter when challenged. The following case studies involve on-site evaluation. Included are the results of a masking agent. On Site Odour Assessment – Sewage Treatment Facility Sampling at this site also involved assessing existing odour control application devices and odour neutralisers along with the formulated odour control solutions by Anotec (masking agent included). The odour source was identified as the inlet works which were exposed. Anotec challenged the assessment by covering the inlet works with a Telecom (now Telstra) canvas workman’s tent. Odour sampling and collection was performed by Anotec and ANSTO. Table 5. Olfactometry Results – On-Site Sewage Treatment Plant Testing Parameter Odour Strength (ODU) Raw Odour 64,560 Raw Odour + Water 11,620 Raw Odour + Odourless 7,940 Raw Odour + Fragranced 10,740 Raw Odour + Masking Agent 52,635

Removal Efficiency (%) 83 88 83 18

Table 6. GC/MS Results – On-Site Sewage Treatment Plant Raw Odour

RO + Water

RO + Odourless

RO + Fragranced

Component (ppb)

RO + Masking Agent

Dimethylsulphide Hydrogensulphide Carbonylsulphide Dimethylsulphide Ethylmercaptan TOTAL

328 9448 39 1779 17 11611

105 1066 17 322 4 1514

5 45 4 29 1 84

134 49 6 109 1 299

268 8569 23 1645 9 10514

Total Odour (ppb)

11611

RE (%) RO + Water

87

Key RE (%) RO + Odourless

99 Key

RO = RE (%) RO + Fragranced

Raw Odour RE (%) RO + Masking Agent

97 9 RE (%) = Removal Efficiency

The results in this case study showed that there is agreement between the olfactometry and the GC/MS results. Olfactometry shows that the Raw Odour is best treated with the odourless formulation, can be treated with water effectively and that the masking agent has little or no effect. The GC/MS results show a similar trend. However, the fragranced formulation seemed to work better than water alone. The masking agent once again did not seem to affect the overall odour by way of reduction. It can be surmised that any fragrant component within an odour control solution will affect the olfactometry. The following case studies do not involve water or masking agents. Based on numerous research results on-site and in the laboratory, the overall outcome is that water’s removal efficiency on various odour emission ranges between 32 and 56%, masking agents removal efficiency is between nil and 17%, fragranced formulations removal efficiency is between 78 – 98%. Water and the odourless formulations do not affect the overall olfactometry or GC/MS results. Fragranced formulations affect the olfactometry result but are not reflected in the GC/MS as, chemically, the removal efficiency is high when assessing these types of products. Masking agents affect the olfactometry result and these results are reflected in the GC/MS data showing little or no change in the overall raw odour concentrations. Based on this research and data, Anotec formulated site specific products for odour abatement using the raw odour samples collected prior to any abatement assessment analysis. On-Site Odour Assessment – Olfactometry and GC/MS results: Plastic Casing Manufacturing Sample collection was via the exit point of the stack leading away from the heat rollers in a plastic casings and sheet facility. Table 7. Olfactometry & GC/MS Results using Plastic Casing formulation Testing Parameter - Olfactometry Odour Strength (ODU) Removal Efficiency (%) Raw Odour 152,450 Raw Odour + Site specific formulation 2631 98 Component Raw Odour (ppb) GC/MS Treated Odour Removal Efficiency (%) Carbondisulphide 3.8 1 72.8 Carbonylsulphide 142 12.3 91.3 Methanol 41,266 497 98.8 Ethanol 96,227 766 99.2 Acetonitrile 2,050 37.3 98.2 Benzene 62,602 546 99.1 Toluene 11,735 336 97.1 mp-xylenes 9,310 117 98.8 Total 2312.6 84 The olfactometry in this case study showed a much more favourable result for removal efficiency using the site specific formulation. However, the overall result suggests that the site specific formulation would succeed in significantly reducing the raw odour from a plastic casings plant. The same procedures were followed through for a variety of industries and odours. It became evident during the research that site specific formulation worked better at significantly reducing odour concentration in various emissions. A trial was run where the formulation used in the Plastic

Casings study was used at a foundry. The results from the combined measurement technique showed removal efficiency between 67% and 72%. A site specific formulation was developed based on raw odour analysis from a ferrous foundry. Odour analysis showed removal efficiency between 89% – 99% CONCLUSION The overall consensus of this paper is that “broad spectrum” odour control, simply put, does not exist. The combined measurement techniques undertaken showed that there is a correlation between the two sets of results, assisting the operator or assessor in determining the best available technology for abatement solutions. By using the combined techniques the operator or assessor will have a complete “picture” of the odorous air stream’s chemistry as well as the human response to it. When applying these techniques to odour abatement technologies, the trial and error method of assessing odours and their most suitable control is no longer applicable and saves valuable time. Assessors and operators will be able to decide on the best course of action for odour abatement by using the combined measurement techniques on various odour abatement technologies on hand. For example, a sewage treatment plant may be considering the use of a biofilter or chemical additive. The combined techniques will show the abatement technique most suitable for the site. In some cases it was noted that, both the aforementioned abatement techniques yielded removal efficiencies >99% and the decision thereon was based on the budget of the operator, one of the treatments yielded better results making the decision instantaneous or, inclusion of one of the treatments was deemed impractical. To use an analogy, if a person is ill with fever they will pay a visit to their general practitioner. The general practitioner will use a thermometer to measure the patient’s temperature. The symptoms surrounding the high temperature or its high temperature reading does not prompt the doctor to write out a prescription. At this stage the doctor may order some blood work, wait for the results and then write out the necessary prescription. In most cases this usually results in the patient’s temperature dropping and the ailment under control. This analogy can be applied to olfactometry. An odour unit or number on a scale will show the assessor or formulator the strength or human response to an odour. Armed with this, the assessor or formulator will have to rely on prior knowledge of the chemistry of the emission based on the site or alleged sources to design and implement a suitable odour abatement solution, hence the trial and error method. GC/MS results of the specific site being assessed will show the exact chemistry of the emission, and, when coupled with the olfactometry will greatly assist in the decision making stage for the best available technology. It will also show the efficacy of any recommended odour abatement products or methodologies or assist in formulating an odour control product prior to commitment and implementation by the facility operator. Over the past twelve years, the results of a combined approach to measurement have consistently shown the formulator an accurate depiction of the effect that a formulated product would have on a malodour. Heavily perfumed or masking agents are not effective for use in a continuous odorous industrial emission but may be a relatively inexpensive treatment for temporary and transient odours in an emergency type situation. Continuous odorous emissions from manufacturing facilities can be effectively treated with site specific formulations based on conducting olfactometry and chemical analysis. As discovered during the course of the research, the use of one product for all situations do not work as the results using sewage odour control at a foundry was not effective. Once the raw odour data was compiled a site specific odour control product was formulated. Odour control products, according to this research, must be formulated as a direct result of the GC/MS and olfactometry analysis. In conclusion the results from this study show that combining olfactometry

and chemical analysis enables the successful formulation and development of suitable odour control technologies.

REFERENCES Air Pollution Control Association (APCA) TT-4-Committee Dynamic Olfactometry Round Robin Interlaboratory Test Results 1977, 1980 AS/NZS 4323.3-2001 Stationary Sources Emissions – Determination of Odour Concentration by Dynamic Olfactometry. Standards Australia, 2001 Australian Standard AS2850-1986 Chemical Analysis – Interlaboratory Test Programs – For Determining Precision of Analytical Method(s) – Guide to the Planning Conduct, Standards Association of Australia, 1986 Dravnieks, A (1976). Development of a Dynamic Forced Choice Method for Measurement of Emission and Ambient Odours. Institute of Illinois Research Project 1976, 96pp. Dravnieks A. et al (1986) Odour thresholds by forced choice dynamic triangle olfactometry: reproducibility and methods of calculation. Journal Air Pollution Control Association. 36(8):900-905.

Duffee R.A. et al (1979). Odors from Stationary and Mobile Sources. Ch.4-8, National Academy of Sciences. O’Brien M.A. (1999). Odour problem: On-site or off-site. Presented at the Water Environment Federation, New Orleans, USA. Stone D.J.M (1994). Chemical fingerprinting of odorous and gaseous emissions. Policy, Regulation and Measurement, Odour Special Interest Group Workshop, Clean Air Society of Australia and New Zealand, Bond University Conference Centre 18 May 1995. Stone D.J.M (1994). What’s that smell in the basement? Quantitative analysis of volatile organic compounds. Applications in odour research. ANSTO, Environmental Science Program. Seminar

Zavras, J. et al (1992). History, Mystery and Chemistry of Odour Control Substances. Presented at the Clean Air Society of Australia and New Zealand Training Activities Committee, Sydney. Zavras, V. (1994). Dynamic Olfactometry – A Forced Choice? Odour Special Interest Group Workshop Perth 27 October 1994. Clean Air Society of Australia and New Zealand. Zavras, V. et al (2002) Environmental Odour Management. Convention Proceedings. 33rd Australian Foundry Institute National Convention 2002.