Standards for the Compressed Air Industry

Filter Integrity LtdA paper presented by Stephen N Smith, Convenor to ISOTC118/SC4/WG1 Compressed air treatment technology, at the Filtration Society 50th Anniversary International Conference and Exhibition, 14th November 2014 at the Riverside Innovation Centre, Chester, UK.

Abstract – Standards for the Compressed Air Industry

Standards for the compressed air industry are broadly divided in to two categories, those for the measurement of compressed air purity specification and compressed air treatment equipment performance. The ISO 8573 series of standards for compressed air purity measurement contains 9 parts of which part 1 enables manufacturers and end users to specify compressed air purity. Compressed air treatment equipment can be validated against ISO 7183 for air drying technology or to ISO12500 Parts 1 through 4 for the removal of oil aerosol, oil vapour, particles and bulk water respectively.

Combined with the vocabulary standard, ISO 3857 part 4 Vocabulary they number in total 16 published standards that have been under development since the mid 1980’s under the management of ISO Technical Committee ISO/TC 118/SC4 “Compressed air treatment technology”. Over the last 18 months the working group has completed a full review of all standards in the series and set priorities for changes required to meet environmental legislation and where appropriate to adopt new methods since they were first published. As a priority the working group has identified that significant changes are required to ISO8573-2:2007 “Test methods for oil aerosol content” due in no small part to recent EU Regulations and availability of suitable reagents to undertake the infrared analysis method detailed. This paper discusses the changes that are being proposed and provides an overview of a timetable for their adoption.

Standards for the Compressed Air Industry – Introduction

Compressed air is a power source that is used worldwide in many industries and manufacturing processes, it is claimed that approximately 70% of all companies use compressed air for some aspect of their operations1. Considered by many to be the 4th utility2 after electricity, gas and water, compressed air is generated on-site and on demand to provide motive power in applications as diverse as snow making machines and microchip fabrication, or beverage manufacture and mine tunnelling equipment. Due to the multitude of applications and diversity of their operating parameters a need for standardisation of the way air purity is expressed and the way in which that purity is determined is required.

ISO 8573 Compressed Air Purity Classes and Purity Measurement Standards Overview

The standards that detail air purity classes and air purity measurement methods are covered by the ISO 8573 series of documents (see Table 1);

NEWS-001Table 1 : Table of standards currently in the ISO8573 series for the specification of compressed air purity and the measurement thereof.

Part 1 of the series was developed to allow both manufacturers and users of compressed air to specify the compressed air purity with respect to the main contaminants experienced in compressed air generation, distribution and application. An overview of these air purity classes is given in Table 2.

NEWS-002Table 2 : Summary table overview of the purity classes of ISO8573-1:2010

The designation principle of the purity class of compressed air at the specified measuring point shall include the following information separated by a colon;

ISO 8573-1:2010 [Particles : Water : Oil]

In some cases users of this system omit to include the year of the document to which they are referring and as there has been a number of revisions since the document was first released the purity statement cannot be relied upon.

A purity class statement relates to the maximum concentration of one of the 3 primary contaminants found in the compressed air. An example of compressed air purity requested by a prospective customer is given as follows;

ISO 8573-1:2010 [1 : 2 : 0]

In this case the customer has requested a purity of Class 1 for Particles, followed by Class 2 which is equal to or drier than -40ºC pressure dew point for Water Vapour and Class 0 Total Oil. In turn the supplier is able to begin to specify equipment to meet the stated requirements given set conditions.

Class 0 is often misinterpreted and does not mean that there is zero contamination. In itself it is not a purity level statement excepting that it is more stringent than the maximum concentration or ‘upper’ limit assigned to Class 1 for the contaminant of interest. Thus a Class 0 statement for Total Oil would have a upper limit of less than 0.01mg/m3 as is the case for Class 1 for this contaminant. It is therefore a statement that needs to be agreed between two parties, and could well be that it aims to match with some other more stringent requirement to control Total Oil. In addition, it is worth noting that if claims are made for a purity level outside of the detection range identified in the relevant standard, then the precise way in which the measurement will be made must be agreed also.

Of further interest is the statement for ‘Total Oil’ which is the combined concentration of oil aerosol and oil vapour. By way of example, coalescing filters that claim to provide air purities of Class 1 can only do so in relation to oil aerosol since they do not impact the oil vapour concentration, or oil in the gaseous phase, in the compressed air. The oil vapour concentration in the compressed air at any one time depends upon the oil type, age, air temperature and compressor running conditions. Thus in this case without a suitable adsorption stage for the oil vapour the class claimed may not be achieved in reality. In the same way, onsite measurements that do not include the vapour content in the total oil statement cannot claim to meet a purity class since the magnitude of the oil vapour component of the statement is unknown.

ISO 12500 & 7183 Equipment Performance Standards Overview
The primary contaminants in compressed air are water, oil and particulates either having been ingested by the compressor from the atmosphere or are as a result of the compression process and the mode of distribution. Equipment to remove these contaminants is available from a broad range of manufacturers all with their own claims regarding performance. As such, in 2007 the first compressed air standards were published detailing test methods for equipment marketed for the removal of liquid and gaseous contaminants. Table 3 provides a list of the standards detailing equipment performance covered by the ISO12500 series for oil aerosol, oil vapour, particles and bulk water. In addition ISO7183 revised in 2007 details compressed air dryer equipment test methods for water vapour removal.

NEWS-003Table 3 : Table of standards currently published for the determination of equipment performance. 

Further additions to the above standards are under consideration including a method that aims to determine the energy consumed whilst operating a compressed air filter over its design life.  This reflects several industry led initiatives to control and reduce energy consumption and thus greenhouse gas emissions for which a standardised means of assessment and rating would be required.

Over the last 10 years the focus of the ISO/TC118/SC4/WG1 working group has been the development of new standards for equipment verification with the publication of ISO12500 parts 1 through to 4, the amendment of ISO 7183 and ISO 8573 parts 1 & 2.  It is now time to review and prioritise any changes required due to new improved techniques and pressure bought to bear by legislation.

New Work Item: ISO8573 Part 2: Test methods for oil aerosol content
The measurement of liquid oil content in its various forms in compressed air is detailed in part 2.  As shown in Table 4, depending upon the oil concentration expected it is possible to select one of two method. Method A which employs coalescing filters and Method B using a membrane that is subsequently analysed by solvent extraction and infrared spectroscopy (see Figure 1).

NEWS-004Table 4: Extract from ISO8573-2:2007 Guide for selection of Test Method.

Fig 1: Oil recovery from the membrane using solvent extraction in accordance with ISO8573:2

Fig 1: Oil recovery from the membrane using solvent extraction in accordance with ISO8573:2

Method B requires that a sample of compressed air is passed through a membrane holder containing 3 layers of high efficiency microfiber glass filtration media and any oil aerosols collected on the membranes are extracted by solvent washing and subsequently by IR analysis.  Since publication initiatives derived from the Montreal Protocol (1987) on substances that deplete the ozone layer3,4 and subsequently under the Kyoto Protocol (1997) to reduce ‘greenhouse gas’ emissions have reduced the availability of suitable reagents. Subsequently the majority of reagents with the transparency required in the alkane C-H stretching region of the infra-red spectrum have been taken out of production and/or their use directly banned by enforcement of international legislation.

One such mandate is European Regulation (EC) No 1005/2009 on substances that deplete the ozone layer and identifies a list of controlled substances in Annex 1 that included such reagents as Tetrachloromethane (or carbon tetrachloride as it is perhaps more commonly known) and Trichlorotrifuoroethane (TCTFE, CFC-113) as having high levels of ozone depleting potential.  Subsequently EU Regulation 291/2011 on essential uses of controlled substances mandated that such substances could no longer be used, for among other things, such purposes as cleaning of components, or for the determination of hydrocarbons, oils and greases in water, air or waste.

Thus without a suitable solvent for IR assays ISO8573-2 Method B is no longer relevant effectively limiting the detection ranges to those of Method A and thus concentrations of oil aerosol in air lower than 1mg/m3 can no longer be measured.  This has further implications when testing coalescing filters for oil aerosol removal in accordance with ISO12500-1 since the majority of equipment performance claims are below that of 1mg/m3. Oil aerosol downstream of the test filter is measured by means of ISO8573-2 and thus without recourse to Method B the scope of ISO12500-1 becomes narrowed.

In lieu of the issues relating to solvent availability for ISO8573-2, Method B a full review and literature search was undertaken by the Working Group throughout 2012/13 which identified two potential candidate methods that would be considered further;

i)             Solvent extraction using Tetrachloroethylene and infrared analysis

Fig 2 : IR transmission spectra of pure Tetrachloroethylene

Fig 2 : IR transmission spectra of pure Tetrachloroethylene

An analysis procedure that applied Tetrachloroethylene (C2Cl4 also known as Tetrachloroethene or Perchloroethylene) in infrared assays of oil and grease in water was published by T.Kaloudis et al. September 20058.  This was followed by application note by Perkin Elmer9, Determination of Oil Content in Membranes Used in Compressed Air Sampling by Infrared Spectroscopy in 2011. Both papers use Tetrachloroethylene solvent for the extraction procedure. Library spectra10 of pure Tetrachloroethylene shows no interference bands in the 3000cm-1 wavenumber region (Fig 2) which would make this reagent seemingly ideal for hydrocarbon analysis.

However to make it practical to store and distribute this reagent stabilisers are commonly added which then introduce a strong absorbance at typically 2875cm-1 wavenumbers (see Fig 3) and interferes with the C-H peak expected at 2860cm-1 wavenumbers for oil.

This would suggest that Tetrachloroethylene is of limited use, however work undertaken by Atlas Copco Airpower, Belgium developed a method where the number of peaks measured is reduced from 3 to 2, and the cuvette pathlength is selected to suit the concentration range.  As a result interference from the stabiliser in this region can be minimised and the accuracy required by the existing standard maintained.

Fig 3 : Typical IR transmission spectra of commercial grade Tetrachloroethylene in quartz glass cuvettes where it can be seen that there is a strong absorbance in the region of 2875cm-1 wavenumbers.

Fig 3 : Typical IR transmission spectra of commercial grade Tetrachloroethylene in quartz glass cuvettes where it can be seen that there is a strong absorbance in the region of 2875cm-1 wavenumbers.

In their proposal the absorbance ‘A’ of the sample is determined from the average for the absorbance of each of the two peaks selected for a given cell pathlength and concentration (equation 1);

A=Avg[A_1:A_2 ] (equation 1)

Where
A is the average absorbance
A1 and A2 are the absorbance of each of the two peaks produced for a given concentration

Absorbance for each peak is determined as follows;

A_n=〖log〗_10 (I_0/I_n ) (equation 2)

Where
An is the absorbance of the selected peak
I0 is the baseline intensity of the selected peak
In is the overall intensity of the same peak

Calibrations have been performed using 1cm and 4cm pathlength cuvettes in the range 60 to 120µg.ml-1 and 1 to 60µg.ml-1 respectively. Analysis by least squares regression resulted in a fit of between 99.74% and 99.96% to that of a straight line corresponding to the Beer-Lambert law for IR transmission through a substance.

Fig 4 : Absorbance spectra obtained from a 4cm pathlength cuvette (left) and 1cm pathlength cuvette (right) with Tetrachloroethylene and compressor oil analytes.  (Note : The absorbance spectra for the 4cm cuvette is shown with the interference peak from Tetrachloroethylene against air overlaid for reference.)

Fig 4 : Absorbance spectra obtained from a 4cm pathlength cuvette (left) and 1cm pathlength cuvette (right) with Tetrachloroethylene and compressor oil analytes. (Note : The absorbance spectra for the 4cm cuvette is shown with the interference peak from Tetrachloroethylene against air overlaid for reference.)

In one example the new proposal developed by Atlas Copco uses an integration of the areas under the curve for two peaks observed here at 2960cm-1 and 2925cm-1 rather than three peaks as detailed currently. It is perhaps worth noting that the exact peak locations are oil specific and as such their location needs to be determined by measurement of the actual oil in the compressed air system. Initial trials conducted by the Working Group are encouraging and if proven to be suitable then this modification to the existing method could be more easily adopted by existing users of the part 1 standard.

Fig 5: Calibration curves obtained for 1cm and 4cm pathlength cuvettes with Tetrachloroethylene and compressor oil analytes.

Fig 5: Calibration curves obtained for 1cm and 4cm pathlength cuvettes with Tetrachloroethylene and compressor oil analytes.

i)             Solvent extraction using n-Hexane and analysis by Gas Chromatography

The determination of hydrocarbon vapours in compressed air is well documented in the method described in ISO8573-5 Test methods for oil vapour and organic solvent content. This document was first published in 2001 and is a modification of the methods described in ISO9486 and ISO9487, workplace air determination of vaporous chlorinated hydrocarbons and vaporous aromatic hydrocarbons respectively.

In place of the above approach which relies upon solvent extraction using carbon disulphide of a sample tube packed with activated carbon, the proposal is to extract collected oil from the membrane using n-hexane and the amount determined by gas chromatography with a flame ionization detector (GC/FID). This approach has found favour with a number of users and has proven reliable for them over the years and if adopted alongside the existing IR method would enable a wider application of the ISO8573-2 standard than at the present time.

Work undertaken by the Institute for Energy and the Environment (IUTA), Duisburg, Germany using a GC-FID has shown that chromatogram can be obtained for the analyte extract obtained from the membrane wash process.  The concentration of oil in solution is proportional to the area under the chromatograph within the limits of n-decane (C10, boiling point = 175 °C) and n-tetracontane (C40, boiling point 525 °C). This area is compared with oil standard solutions (synthesized with reference oil taken from the compressor before measurements if possible) and the amount of oil is calculated with regard to a regression graph produced during calibration.

The proposed solute for this method is n-hexane which is known to be available widely, be of good purity and relatively inexpensive. Alternatives such as
n-pentane and n-heptane would also be suitable but due to the diverse nature of compressor coolants and lubricants it is good practice to investigate the suitability of the chosen reagent with the oil of interest regarding solvency.

Calibration of the GC-FID response for the oil of interest is conducted having produced 10ml each of calibration standards in the range 20µg/ml to 300µg/ml along with a blank n-hexane sample.  Where several oils are known to be present then the response for these is determined separately.  If the source of the oil cannot be identified then a ‘reference’ oil can be used to which all results are stated as being equivalent to.  By analysis the area under the chromatogram obtained from the GC-FID between C10 and C40 is computed.

Fig 6 : Typical chromatogram for compressor oil after elution with n-hexane.  The C10 and C40 markers can be clearly seen.

Fig 6 : Typical chromatogram for compressor oil after elution with n-hexane. The C10 and C40 markers can be clearly seen.

A plot of Peak Area for an average of 3 repeat samples against the 10 individual aliquots is produced, an example of which can be seen in Fig7.

Fig 7 GC-FID Peak Area response for calibration oil samples in the range 20µg/ml to 300µg/ml in n-hexane. Detection down to 4µg/ml has been demonstrated.

Fig 7 GC-FID Peak Area response for calibration oil samples in the range 20µg/ml to 300µg/ml in n-hexane. Detection down to 4µg/ml has been demonstrated.

In this case by least squares regression a fit of 99.93% was achieved to that of a straight line and demonstrates the linearity of the technique over a broader range of concentrations than is currently possible using IR.

Once the total mass of oil extracted from the membrane has been determined by either of the proposed methods the concentration of oil in the compressed air can be calculated.

Project planning
Having now been accepted as a New Work Item Proposal (NWIP) by ISO/TC118/SC4 further work will identify the efficacy of these proposals over a 2 year project plan has been established (see Fig 8) set within the confines of the ISO process for the revision of a standard.

Fig 8: ISO 2 year project plan

Fig 8: ISO 2 year project plan

To establish the repeatability and reproducibility of both proposed methods the working group agreed at the October 2014 meeting to conduct an exchange of samples using one oil type at three levels of concentration when applied to membrane discs by one laboratory.  Results from the test program will shared between the members prior to the next working group meeting scheduled for March 2015.  At this time a Draft International Standard (DIS) will be ready for release.

Summary – Standards for the Compressed Air Industry

Under the guidance and control of ISO working group ISO/TC 118/SC4/WG1 there are now a total of 16 published standards that have been in circulation commencing in the mid-1980’s.  The drive up until now has been on the generation of new standards initially for the measurement of compressed air purity but in more recent years for compressed air equipment performance validation.  A review of the full suite of documents throughout 2012/13 revealed that a number of users were experiencing difficulties complying namely with ISO8573-2 Oil Aerosols.

Due to the Montreal and Kyto protocols restrictions on the use of controlled substances to protect the ozone and reduce global warming have come into force.  More recently these initiatives have resulted in local regulation such as that of EU Regulation 291/2011 preventing the use of such reagents for oil analysis.  This has direct implications when working in accordance with ISO8573-2 when measuring oil aerosol concentrations below that of 1mg/m3.

A modification to the existing method has been reviewed to enable solvent extraction with Tetrachloroethylene and in addition a new method centred on the use of GC/FID instrumentation and membrane extraction with n-hexane proposed.  Initial work suggests that both new methods have been found to provide the necessary level of linearity across the range of concentrations of interest.  The next step is to widen the trials to include 5 participating laboratories to establish repeatability and reproducibility of both proposal.  The results of the lab trials are due to be available before the end of November 2014 at which point a decision on whether to expand the trials will be taken.  A two year project plan has been developed with the aim of publishing a new version of the part 2 standard for oil aerosol measurement before the end of 2016.

References – Standards for the Compressed Air Industry

1 Vane or Screw Compressors, Plant and Works magazine, 9th July 2014 available. Click Here >

2 The fourth utility, Maintenance Matters magazine, March 2006 edition, available from the British Compressed Air Society, London. Click Here >

3 United Nations Treaty Collection under the Vienna Convention for the Protection of the Ozone Layer, Vienna 22 March 1985, Chapter XXVII Environment, Section 2.  Available from United Nations Publications or online. Click Here >

4 United Nations Treaty Collection under the Montreal Protocol on substances that deplete the ozone layer, Montreal 16 September 1987, Chapter XXVII Environment, Section 2a.  Available from United Nations Publications or online. Click Here >

5 Kyoto Protocol to the United Nations Framework Convention on Climate Change, Dec. 10, 1997, U.N. Doc FCCC/CP/1997/7/Add.1, 37 I.L.M. 22 (1998).

6 EC Regulation No 1005/2009 of 16 September 2009 on substances that deplete the ozone layer. Available from the Official Journal of the European Union, EUR-Lex Access to European Law.  Click Here >

7 EC Regulation No 291/2011 of 24 March 2011 on essential uses of controlled substances other than hydrochlorofluorocarbons for laboratory and analytical purposes in the Union under Regulation (EC) No 1005/2009 of the European Parliament and of the Council on substances that deplete the ozone layer.  Available from the Official Journal of the European Union, EUR-Lex Access to European Law. Click Here >

8 Validation of an FT-IR Method for the Determination of Oils and Grease in Water, with the use of Tetrachloroethylene as the Extraction Solvent.  T.Kaloudis et al. Proceedings of the 9th International Conference on Environmental Science and Technology, Rhodes Island, Greece, 1-3 September 2005.

9 Determination of Oil Content in Membranes Used in Compressed Air Sampling by Infrared Spectroscopy. Application Note; Author Avdhut L. Maldikar, Ph.D. Perkin Elmer, Pvt. Ltd., Kasarvadavali, Thane (West), India. Copyright Perkin Elmer 2011. Click Here >

10 National Institute of Standards and Technology, NIST Chemistry WebBook Standard Reference Database Number 69.  Tetrachloroethylene C2Cl4
 Click Here >

For more information about Standards for the Compressed Air Industry contact :

Filter Integrity Ltd
NETPark Incubator
Thomas Wright Way
Sedgefield
County Durham
TS21 3FD
UK

Contact this Company

Subject (required)

Name (required)

Company Name (required)

Company Activity (required)

Telephone Number

Your Email (required)

Your Message (required)

[recaptcha recaptcha-174]

Comments are closed.