Cleanroom Wiper Applications for Removal of Surface Contamination

Chapter 14

Cleanroom Wiper Applications for Removal of Surface Contamination Jay Postlewaite, Brad Lyon and Sandeep Kalelkar Texwipe, an ITW Company, Kernersville, NC, USA

Chapter Outline 1 Principles of Wiping for Removal of Contaminants 550 1.1 Why Wipe? 550 1.2 What is the Contamination? 550 1.3 Why Wiping Works 551 1.4 How to Wipe 552 1.5 Wiping Methods 553 2 Types of Wipers 555 2.1 Glossary of Terms 555 2.2 Knit Synthetic Wipers 557 2.3 Microdenier/Microfiber Wipers 558 2.4 Woven Fabric Wipers 559 2.5 Nonwoven Fabric Wipers 560 2.6 Foam Wipers 562 2.7 Selecting a Cleanroom Wiper 563 3 Wiper Testing 564 4 Methods to Assess Wiper Quality 565 4.1 Particles and Fibers 566 4.2 Ions 575 4.3 Extractable Matter 576

4.4 Using Box and Whisker Charts to Evaluate Wiper Consistency as the Measure of Quality 576 4.5 Advantages and Disadvantages of Wiper Test Methods 582 5 The importance of Automation 583 5.1 Wiper Edge Treatments 583 5.2 Automated Manufacture of Wipers 583 6 Applications 585 6.1 Semiconductors 585 6.2 Disk Drives 586 6.3 Pharmaceuticals 586 6.4 Biologics 586 6.5 Medical Devices 587 7 Current Trends in Wiper Technology 587 8 Future Developments in Cleanroom Wipers 588 References 589

Developments in Surface Contamination and Cleaning, Volume 11. https://doi.org/10.1016/B978-0-12-815577-6.00014-1 Copyright © 2013 Elsevier Inc. All rights reserved.

549

550 Developments in Surface Contamination and Cleaning

1 PRINCIPLES OF WIPING FOR REMOVAL OF CONTAMINANTS 1.1 Why Wipe? The growth in the technology and manufacturing of high-end products, such as computer chips, medical devices, and pharmaceutical drugs in the microelectronics and life sciences sector over the past decade, has been significant. These industries typically utilize extremely clean environments for the manufacture of these products. Cleanrooms are constructed in semiconductor fabrication facilities to minimize the negative impact that environmental contamination can have on product yields. In the pharmaceutical, biologics, and medical device industries, environmental cleanliness is essential to minimize the safety risk of the product to patients receiving the medication or device. Increasingly, as drugs get more potent, the risk of residual contamination and crosscontamination is even greater requiring considerations of protecting the personnel in the manufacturing area through contamination control. Therefore, manufacturing companies expend significant resources constructing cleanroom environments, air handling, and monitoring systems to ensure that airborne contamination levels remain in control at the International Standards Organization (ISO) class level desired, and specified, in operation. While this is a necessary and critically important step in minimizing contamination risk to products and people, it is not sufficient. Within an operational cleanroom, measures and protocols must be established to ensure that surfaces, equipment, instrumentation and, most importantly, people do not inadvertently contribute contamination burden to the environment, thereby risking the product or personnel. Physical methods of contamination removal, therefore, play a critical role in minimizing contamination risk. Wiping surfaces is one of the most effective methods to physically remove contamination. While one may use other methods such as vacuuming, blowing compressed gas, and irrigation, these methods tend to have side effects that may further compromise the environmental contamination in the room. Wiping is most effective because it serves to physically capture and hold in place residues without spreading them further. It is also important to note that there are a variety of surfaces to be wiped, in various shapes, sizes and materials, and in rooms that are certified to be operational at various ISO classes. These will often require wipers of different kinds that are particularly suited to a given surface or contamination challenge. In this context, it is important to recognize that swabs and mops effectively serve to wipe contamination off surfaces as well. They are typically comprised of cleanroom wiper materials that contact surfaces both large and small to ensure that the act of cleaning does not inadvertently introduce any residues into the manufacturing areas.

1.2 What is the Contamination? One of the most important principles to recognize to understand wiping in cleanrooms is that the contamination is typically of the invisible kind. While

Cleanroom Wiper Applications Chapter

Volatile organic contaminants

14

551

Airborne molecular contaminants

Nonvolatile Microorganisms residue Adsorbed Ions molecules

Particles Metals

Fibers

Surface FIGURE 14.1 Artist illustration of different kinds of surface contamination.

some macro residues may be visible to the naked eye, cleanroom contamination is typically concerned with micro-level residues, often extending from below 0.1 μm to above 100 μm. Even with the superior air handling and filtration offered by proper cleanroom construction, surfaces can be contaminated with a variety of different residues and materials that differ physically and chemically. Small submicrometer (micron) particles, larger visible filamentous fibers, acids, bases, salts, organic matter, a variety of materials both solid and liquid form used in the process, and nonvolatile residues (NVRs) are all examples of surface contaminants that may be commonly present and need to be removed. This is demonstrated in Fig. 14.1. The risk with these contaminants is not merely that they will exist in a cleanroom environment, potentially risking the product integrity, but that they may migrate to other locations in the room if they are not promptly, frequently, and periodically removed. In fact, using methods other than wiping for removal can increase the potential of such undesirable migration. Therefore, most cleanroom operations will document specific cleaning procedures, tools and frequencies as part of their quality assurance systems to ensure that product integrity will not be compromised.

1.3

Why Wiping Works

To understand the principles behind the effectiveness of wiping to remove contamination from surfaces, it is necessary to understand the adhesion of contamination to surfaces. For a discussion of adhesion forces between contaminant particles, the reader should refer to the review by Bowling [1]. The physical action of wiping provides about 0.689 MPa (100 psi) of pressure on the surface. Capillary forces dominate in the adhesion of contaminants to surfaces. Breaking up or lowering this capillary adhesion is critical to the removal of soils. In principle, the goal is to use materials and wipers that will result in the contaminants having greater attraction for the wiper than the surface from which they must be

552 Developments in Surface Contamination and Cleaning

removed. The most effective way of lowering this capillary adhesion factor is to use liquids such as water, or other solvents with lower surface tension such as alcohols or other cleaning agents. Therefore, in general wipers that have been wetted with a cleaning solution are more effective at removing contamination from a surface. This is particularly true when electrostatic interaction is a factor in the adhesion of contamination to a surface. A prewet wiper would effectively neutralize the electrostatic effect.

1.4 How to Wipe It is important to recognize that wiping surfaces in a cleanroom environment follows a prescribed technique that is designed to minimize spreading of the contamination and for the greatest efficiency of the wiping protocol. The technique calls for a systematic cleaning of all surfaces from the cleanest to the dirtiest areas. For example, even within a large cleanroom, one would begin with areas furthest from the entrance towards the ones nearest it. Likewise, one would begin cleaning from the top of contaminated surfaces towards the bottom asthe lower levels would be expected to have greater contamination. A wiper is used quarter-folded as shown in Fig. 14.2a–c, thus providing four usable surfaces on each side of the wiper. This is done is order to ensure that one wiping surface on which the contamination is once picked up is not re-used, which can risk releasing the contamination back to the surface. It is also important to be watchful to ensure that the physical act of turning the wiper over to produce a fresh unused wiping surface does not release contamination. The actual wiping technique is to use consistent pressure using the fingers and palm of the hand to cover the area to be cleaned with firm overlapping strokes as illustrated in Fig. 14.3. This ensures that the entire surface is covered. A similar technique is used when wiping large surfaces with mops, which are effectively just larger wipers. In some cases, such as isolators, smaller mops may be used and wipers may be merely the contact point that the mop hardware uses to wipe a surface. Likewise, swabs may be used to wipe contamination from narrow channels,

FIGURE 14.2 This figure demonstrates how a wiper is folded in mid-air to deliver the quarterfolded wiping surface.

Cleanroom Wiper Applications Chapter

14

553

FIGURE 14.3 This figure demonstrates how the wiping is performed on a surface using firm overlapping strokes.

edges and corners of hard to reach surfaces, and they are usually made of cleanroom wiper materials as well.

1.5

Wiping Methods

Wiping can be thought of as a way of cleaning, whether it is the removal of a spill, the control of spills, the removal of surface contaminants, or the application of cleaning solutions. The type of wiper material used for the various applications is dependent upon the environment, the intended need, and economic considerations. The use of dry wipers can be broken down into four general categories: spill control and removal, cleaning solution application, a work surface pad/ protective covering, or cleaning. There are many different wiper designs to meet the various needs of these categories. A. Spill control and removal Wipers used for spill control and removal usually have the greatest sorption capacity so that the fewest number of wipers can be used. Wipers incorporating some sort of natural fiber will leave a surface visibly dry as the liquid is captured within the complete fiber, while synthetic-based wipers will leave a thin liquid layer on the wiped surface. Synthetic fibers will adsorb liquid onto the surface of the fiber and between multiple fibers (not completely through the fiber as natural fibers do) and cannot compete as well for removing a liquid, thus leaving a thin film on the surface that is wiped. Consideration of the work environment is very important when evaluating the proper wiper for spill control and removal. Not only does the ISO class environment need to be considered but so does the surface being wiped. Some wipers are designed for more abrasive surface textures but these may have less sorptive capacity, while wipers with higher capacity may shed more particles when used on an abrasive surface.

554 Developments in Surface Contamination and Cleaning

B. Solution application and/or removal Many industries require the application of unique liquid solutions such as cleaning agents, disinfectants, or specialty chemicals. These solutions may be sprayed or poured on the surface then evenly dispersed with a wiper; or the dry wiper may have the solution applied at time of use and then used for wiping. The selection process for the appropriate wiper includes the cleanliness, abrasion resistance characteristics, sorptive capacity, and chemical compatibility. Quite often the solution applied has a defined surface contact time before it is removed which may allow the solution to dry and require extra force for removal, requiring different wiper characteristics than those considered for application. An easy analogy to this process is applying and removing car wax. A sponge is used to apply the paste of liquid wax, but a soft, clean material with a large surface area is used for efficient wax removal without scratching the painted surface. C. Work surface pad/protective covering Some materials may have a very sensitive surface and, in such applications, a wiper is used as an underlayment to protect the object from a hard work surface which may create damage. Depending upon the object and the environment being worked in, cleanliness and material thickness combined with the density are the main attributes for consideration. A very clean wiper may also be used to wrap parts as part of packaging to protect the parts after final assembly and cleaning. D. Cleaning Dry wipers alone are usually not used to wipe or clean a dry surface (microdenier fabrics may be an exception) since there are many physical forces in play which hinder the effective capture and removal of contaminents. But there are instances when wiper materials are made as a continuous roll. In these applications, the product is usually being unwound from one roll and rewound after being introduced to a surface in an automated fashion for surface cleaning, polishing, or even burnishing. The roll is indexed incrementally as each new part comes in contact with the web exposing a clean spot for the operation. Wiping a surface with a dry wiper generally is avoided due to the fact that a dampened wiper is more effective at removing contamination from the surface than a dry wiper. A dry wiper may simply push the contamination around as the electrostatic attraction between a particle and the surface may be greater than the attraction force between the particle and the wiper. (An exception to this is the use of microdenier wipers that have an increased surface area and may capture and remove bacteria from a surface). The use of dampened or pre-wet wipers provides an optimum approach to removing surface contamination. The liquid used to wet the wiper can provide a means to lower the surface tension between particles and the surface (by lowering adhesion forces) to aid in particle removal. Once the contaminants are freed, they become trapped in the wiper, thereby avoiding

Cleanroom Wiper Applications Chapter

14

555

recontamination of adjacent areas. It is important to realize that there is an optimal range of wetness for a pre-wetted wiper for general cleaning applications. If a wiper is too wet, not only is excessive solution left behind on the surface that may possibly contaminate the area being cleaned, but the contaminants being removed are simply pushed around the surface entrapped in the solution instead of being transferred onto the wiper. Likewise, if a wiper is undersaturated, the wiper will be less effective in removing contaminants. It is for these very reasons that it is better to use a manufactured pre-wet wiper compared with to a pre-wet wiper made at time of use which will have varying amounts of wetness based upon the person applying the liquid.

2

TYPES OF WIPERS

Cleanroom wipers and methods to use and evaluate such products have evolved in parallel to the advances in cleanroom and manufacturing needs. In general terms, cleanroom wipers are broken down into four broad categories which reflect the raw material manufacturing method. The four categories are: Knit synthetic, Woven, Non-woven, and Foam. Knit, woven, and nonwoven materials are considered traditional textile materials. However, before examining the details of wiper construction techniques, consideration should be given to the various attributes of wiper performance and a glossary of terms and definitions used to describe cleanroom wipers is included below for reference.

2.1

Glossary of terms

Particles/fibers: one method of evaluating wiper performance. Not all wipers perform the same and different environments and processes have differing needs Ions: generally more important in microelectronics, semiconductor, and aerospace applications NVRs (nonvolatile residues): an important consideration, in particular, if product is used with a solvent which could leach residue from the wiper and contaminate a surface Sorption: characteristics: how fast and how much liquid a material will hold Chemical resistance/interaction: will the wiper withstand the chemicals that may be encountered or will the wiper breakdown and create contamination? Raw material components: impacts cleanliness, cost, overall performance, and chemical interactions Metals: are there metals remaining from the manufacturing process either on the surface or integrated into the fiber that can be extracted under certain conditions that could be a contaminant? Leave behind: anything remaining after wiping a surface, may or may not be visible to the naked eye

556 Developments in Surface Contamination and Cleaning

Extractable: something removed by mechanical or chemical force that may be considered a contaminant ESD: electrostatic discharge Abrasion characteristics: does the wiper create particle contamination when used on rough surfaces? Packaging considerations: ease of use for the operator and proper amount contained in bag so that excess wipers are not disposed of before use. This is of particular concern with sterile products. Silicone: depending upon the industry, it can either be a positive manufacturing aid (for example, medical devices) or a major concern creating possible defects (electronics and painting). Most wipers are manufactured and tested to be silicone free. Outgassing: is the release of a gas that was dissolved, trapped, frozen or absorbed in some material. Outgassing is of particular concern when trying to maintain a clean high vacuum environment, or in critical environments concerned about airborne molecular contamination (AMC). Practical application: cleaning, spill control, environment protection, packaging, solution application Microbial considerations: generally thought of as contaminants in the life science industry and the impact if a microbe is transferred on or into a product that ultimately may be inside the body, but should also be considered as a particulate contaminant in other industries Sterile/non-sterile: depends on the manufacturing environment and the product being made. Most sterile wiper products are based on a non-sterile product with similar performance attributes. Common terms/Definitions relative to cleanroom wipers: Yarn: A generic term for a continuous strand of textile fibers, filaments, or material in a form suitable for knitting, weaving, or otherwise intertwining to form a textile fabric. Filament: A fiber of an indefinite length such as found naturally in silk. Manufactured fibers are extruded into filaments that are converted into filament yarn. Filament yarn: A yarn composed of continuous filaments. Denier: A direct numbering system for filament yarns. Lower numbers represent a finer size and the higher numbers represent coarser sizes. Denier is a weight-per-unit length measure of a linear material. This is numerically equal to the weight in grams of 9000 meters of the material. Denier per Filament (dpf): The denier of an individual continuous filament. In filament yarns, it is the yarn denier divided by the number of filaments. Yarn Denier: The denier of a filament yarn. It is the denier per filament and the number of filaments in the yarn. A common yarn size for cleanroom wipers is 1/70/34 (1 ply, 70 denier, 34 filaments). Ply: The number of singles yarns twisted together to form a plied yarn.

Cleanroom Wiper Applications Chapter

14

557

Spun yarn: A yarn consisting of staple fibers usually bonded by twist. Spun yarn may be made of either natural fibers or filaments cut to a specified length. Staple fiber is used in the textile industry to distinguish between natural or cut-to-length manufactured fibers from filaments. Since spun yarn is made from multiple short fibers which may shed more particulates than filament yarn, it is commonly not used in cleanroom products unless there is no other option. Knit: A method of manufacturing a textile fabric by interconnecting loops of yarn. Woven fabric (weaving): A method of manufacturing a textile by interlacing two yarns to cross each other at right angles. The yarn that runs the length of the fabric is called the warp and the yarn across the fabric is the weft or filling. Nonwoven fabric: An assembly of textile fibers held together by mechanical interlocking in a random web. The web may be held together by fusing (thermoplastic fibers), bonding with a chemical additive, or hydro-entangling in which high pressure water jets are used to entangle the web materials. Foam: A substance that is formed by trapping pockets of gas in a liquid or solid. For discussions concerning cleanroom wipers, there are two primary types of foams: open-cell structured foams (also known as reticulated foams) and closed-cell foams.

2.2

Knit synthetic wipers

Knit synthetic wipers are the material of choice for the most critical cleaning applications. A knit material is manufactured by interlocking loops of yarn as shown in Fig. 14.4 [2]. For critical cleaning applications the yarn of choice is filament polyester. Filament nylon may also be used. A filament yarn is extruded in a continuous form which provides a stronger yarn that has better abrasion resistance and thus less particle generation. Polyester has replaced nylon as the yarn of choice because it is cleaner (relative to extractable matter and particle generation), has better chemical resistance, and is more economical as a raw material. Knit fabrics as a base wiper material also present the most versatility related to product development. Knitting machines have relatively easy to change design parameters with the ability to process small amounts of product. This may enable product development to occur in a relatively short and economical setting, whereas other manufacturing methods require much larger production runs and encounter more difficulty in changing design parameters quickly. Knit fabric from a design/performance perspective using synthetic yarns allows for many characteristic variations. For instance, if the wiper is to be used in an environment that may have rough surfaces, a very tight fabric structure with minimal stretch is preferred as this is less likely to snag and break the yarn releasing extra particulate matter. If the objective

558 Developments in Surface Contamination and Cleaning

FIGURE 14.4 An illustration of the interlocking loops of yarn formed during the knitting process.

property is sorptive capacity, a looser knit which has more stretch may be the basis for the design. The type of yarn used may also influence the design and performance characteristics. Just as there are many different types of yarns used for apparel applications which influence the hand (feel), moisture management characteristics (performance apparel which transfers moisture from the skin to the outside of the garment allowing for faster evaporation) to microfiber/microdenier materials (very fine yarn, like silk), the same yarns can be used for cleanroom wipers providing different performance characteristics. (A separate discussion of microdenier fabrics follows below). As with most characteristics, a favorable change of one performance characteristics often has a corresponding detrimental change. For example, microdenier yarns provide superior wiping characteristics, but the downside is that the yarn to make the wipers is much more expensive than standard yarn, thus increasing the cost of the wiper. The limitation of knit wiper design and performance is really the imagination of the fabric designer combined with the processing capabilities of the manufacturer and obviously determining what the market will bear relative to cost.

2.3 Microdenier/Microfiber Wipers Microdenier or microfiber (these terms are often used interchangeably) materials have very unique wiper properties, especially considering the material is a synthetic based product. Microdenier materials have been used in consumer applications for many years but are just starting to receive attention in controlled environments. The favorable attribute of microdenier materials is

Cleanroom Wiper Applications Chapter

14

559

the ability to clean a smooth surface such as glass or stainless steel without having to use a solution such as isopropyl alcohol (IPA) or deionized water (DIW). A practical demonstration is dry wiping a pair of glasses or a smart phone display – fingerprints and skin oils are removed without scratching the surface. The very fine filament fibers make the microdenier behave in this fashion, but the very small size of fibers also weakens the fibers making the fabric more susceptible to abrasive surfaces and broken filaments. The broken filaments may also release many more particles and fibers into the work environment. Starting in the late 1990s a number of studies were conducted comparing microdenier materials used for mopping and wiping surfaces in health care environments with the cleaning efficiency of standard accepted materials [3]. In short, it was discovered that the properties of microdenier materials are more effective at removing bacteria from hard surfaces including old laminate surfaces with microfissures in which bacteria may collect. The bacteria tested included Methicillin-resistant Staphylococcus Aureus (MRSA) and Clostridium difficile. In most cases, cleaning the surface with a microdenier material wetted with deionized water resulted in almost complete removal of culturable bacteria. The comparison wiper is known as the J Cloth, a common and accepted wiper used in food service and health care industries.

2.4

Woven Fabric Wipers

Woven fabric was used for one of the original products specifically designed for cleanroom applications, cotton twill produced by The Texwipe Company in 1964. Woven fabrics in general terms are two yarns that cross each other at right angles. Unlike knit fabrics that allow for multiple spools of yarn to be loaded on a single machine and subsequently produce a material that can be converted into a wiper after further processing, woven material has many preceding manufacturing steps (creeling, warping, slashing, and drawing in) before fabric formation (weaving) even starts and generally is on a much larger production scale. In general terms, woven materials are the slowest and most complex to make due to the manufacturing process steps involved and is thus generally the most expensive. Woven fabrics such as knit fabrics have many design options, but are more limited by the type of equipment a weaver may have. Most weaving plants have a specialty and do not have all the various types of looms to make all the different fabric constructions, whereas a knitting machine has much more versatility and can make many different options relatively easily. Generally speaking, woven fabrics have more rigidity (minimal stretch) and are a denser fabric than knits. These characteristics are good if there is a concern about abrasion but may limit the sorption capacity of the material. If the fabric construction is very dense (many woven fabrics are), it may also be much more difficult to clean when compared to knit fabrics.

560 Developments in Surface Contamination and Cleaning

Plain weave

5-Harness satin weave

2/2 Twill weave

3/3 Twill weave

FIGURE 14.5 Common patterns found in woven materials.

Both knitted and woven fabrics must be cleaned after the basic fabric formation is complete. During the yarn and fabric manufacturing steps, lubrication aids (waxes or mineral oils) must be applied. The first step to producing a cleanroom compatible wiper is to remove the lubrication aids and gross contamination. This process step is referred to as scouring. The scouring process and chemicals are specially designed for materials used in cleanrooms compared to standard textile applications. Fig. 14.5 represents four common woven constructions. The 2
2 twill weave is very common for cleanroom wiper constructions. The difference between the four fabric styles is the number of yarn ends covered by another yarn. For example, the plain weave is simply alternating an over under pattern of the yarn one at a time. By contrast, the 2
2 twill has the yarn end on top of two ends followed by going under two ends. This alternates throughout the pattern.

2.5 Nonwoven fabric wipers Nonwoven fabrics are broadly defined as a structure consisting of fibers which are mechanically bonded together in a random pattern (knit and woven fabrics have a defined structure/pattern) as illustrated in the scanning

Cleanroom Wiper Applications Chapter

14

561

FIGURE 14.6 SEM micrograph of a hydroentangled nonwoven fabric illustrates the random fiber pattern of the construction.

electron microscope (SEM) micrograph in Fig. 14.6. They are not made using conventional textile systems with a warp and weft and do not require conversion of the fibers to yarn. Nonwoven fabrics are engineered structures that offer high speed, large volume production, and thus are the most economical material when compared to traditional cleanroom fabrics. The cost advantage achieved by using nonwoven materials is also a disadvantage when comparing knit and woven materials for design considerations: the materials are offshoots of commercialized products for other applications often for the consumer market. Due to this reason, one of the historical limitations of nonwoven materials is the relative lack of cleanliness compared to other materials. Nonwoven materials are not further processed relative to cleaning after the fabric manufacturing step, and thus the particle counts and ions values are greater than products that have further processing to reduce contaminants such as laundered, knit polyester wipers. Most nonwoven products are not recommended for use in ISO Class 4 or cleaner environments due to this reason. Many nonwoven cleanroom wipers contain some sort of cellulosic fiber as part of the product. The cellulosic base offers a material that easily removes spills, enhances sorptive capacity, and wipes a surface dry like a paper towel does in a home scenario. Cleanroom nonwovens are not limited to cellulose based materials. Polymer-based fibers such as polypropylene are also used. Many times, polymer-based nonwovens are manufactured by a process known as spun-bond. This process involves the direct extrusion of pellets to fiber which, in turn, is directly converted into a fabric web. The advantage of this

562 Developments in Surface Contamination and Cleaning

type of manufacturing is the cleanliness of continuous filament fibers with minimal processing aids added. This type of material is generally the cleanest of the nonwoven materials but not as clean as a laundered polyester. Very often these fabrics are used for their special chemical resistance properties or as the base material for alcohol wetted wipers. As a dry wiper, the pure polymer will be hydrophobic, but additives can be added to make it hydrophilic. Unfortunately, often the additives used may negatively impact other cleanliness characteristics such as ions or non-volatile residues.

2.6 Foam Wipers Unlike traditional wiper matrials that use fibers that physically interlock in some method to capture liquids, foam is formed by capturing gas into a polymer matrix as it forms by chemical reaction. Fig. 14.7 illustrates the structure of an open-cell foam and Fig. 14.8 shows the structure of a closed-cell foam. The gas can be entrained into the polymer or it can be formed in situ. Foam has the advantage of capturing and holding a large amount of a liquid substance without having much mass. Foam may also release liquids in an even amount. An easy analogy is comparing the application of car wax using a foam pad to a traditional textile such as a towel. The foam pad holds more wax per given area and releases back onto the car surface in a more even amount for a longer period of time for the same amount of wax applied by the applicator.

FIGURE 14.7 SEM micrograph of an open-cell foam structure.

Cleanroom Wiper Applications Chapter

14

563

FIGURE 14.8 A photomicrograph of closed-cell polyurethane foam structure at 10X magnification.

The negative of using foam is that it may not be sufficiently clean for critical applications even if it is laundered. Traditionally, just because of the manufacturing process and product design, foam materials have more particles than traditional cleanroom wiper materials, such as laundered synthetic wipers, but foams may be cleaner than some nonwovens. Also, foams carry more unwanted contamination due to their formulation, especially ionic and NVR levels compared to traditional textile-based wipers. The biggest advantage of foam – liquid retention – also makes it difficult to clean in that the medium used to wash and the very contaminants meant for removal are held inside the body of the foam.

2.7

Selecting a Cleanroom Wiper

In selecting a cleanroom wiper for use in any cleanroom application, it is always essential to understand the cleanliness level of the background environment and the nature of the residues, soils and solutions to be either removed or applied. Table 14.1 serves as a guide to the selection of a cleanroom wiper. It should be noted that currently there is no established test to determine if a particular wiper is suitable for an environment; the table represents recommendations based upon historical use of the general types of wipers in specific environments. The user should always make the final decision if the wiper meets the needs of the application.

564 Developments in Surface Contamination and Cleaning

TABLE 14.1 A Guide for Selecting the Optimal Wiper Substrate for a Cleanroom Class Cleanroom Class

Optimal Substrate

Comments

ISO Class 3–4

Polyester knit, sealed border or sealed edge

Lowest particle, fiber, NVR, and ionic levels. Wipers are sealed, laundered, and cleanroom bagged.

ISO Class 4–5

Polyester knit, unsealed edge

Low particle, fiber, NVR, and ionic levels. Particle and fiber levels are higher than in sealed border or sealed edge wipers. Wipers are cut, laundered, and cleanroom bagged

ISO Class 5–7

Non-woven materials

Can be used in areas in which moderate levels of contamination control is required.

Composite materials

Can be used in situations where some particle control is needed. High absorbency.

Cotton

Use where heat resistance and slight abrasion characteristics are required

Polyurethane foam

Use in situations where some particle control and good absorbency are needed.

> ISO Class 7

3 WIPER TESTING Wipers are part of the protocol used in the control of contamination in a cleanroom. They are used to clean hard surfaces, equipment, chambers, and tools, clean up spills, or serve as a work surface. As part of cleaning protocol and functioning of a cleanroom, evaluation of wipers for their cleanliness is important. Cleanrooms are classed as to the level of particle contamination in the air. Table 14.2 below defines the number of airborne particles per cubic meter allowed for each cleanroom class. Wipers have different levels of contamination depending on the type of material that comprises the wiper (natural materials yield higher levels of particles than man-made materials), the construction of the wiper (nonwoven, woven, or knit), and the cleaning method (none, cleanroom laundry, or automation). The cleanliness of a wiper used in a cleanroom needs to match the level of the cleanroom. To know this, the cleanliness level of a wiper must be evaluated. Depending on the user, different contamination characteristics are important to the functional operation of their cleanroom. Different wiper performance characteristics are important depending on its use. It is critical that a wiper is evaluated for its fitness before it is used in a cleanroom. It must fit its use and the contamination class of the cleanroom in which it is being used.

Cleanroom Wiper Applications Chapter

565

14

TABLE 14.2 ISO 14644-1 Cleanroom Classification by Particle Contamination Level [4] Maximum Number of Airborne Particles Allowed per Cubic Meter Class

≥0.1 μm

ISO 1

10

2.37

1.02

0.35

0.083

0.0029

ISO 2

100

23.7

10.2

3.5

0.83

0.029

ISO 3

1000

237

102

35

8.3

0.29

ISO 4

10 000

2370

1020

352

83

2.9

ISO 5

100 000

23 700

10 200

3520

832

29

ISO 6

6

237 000

102 000

35 200

8320

293

7

6

1 020 000

352 000

83 200

2930

7

7

3 520 000

832 000

29 300

8

35 200 000

8 320 000

293 000

ISO 7 ISO 8 ISO 9

1.0
10

1.0
10

8

1.0
10

9

1.0
10

≥0.2 μm

2.37
10 2.37
10

8

2.37
10

≥0.3 μm

1.02
10 1.02
10

≥0.5 μm

≥1 μm

≥5 μm

Cleanrooms are designed to control contamination either by managing contamination to low levels within the environment, or more preferably by keeping contamination outside the controlled environment to begin with. Different industrial areas may be variably sensitive to different sources of contamination in their operations. A pharmaceutical company may consider large fibers to present a significant risk to their aseptic process for a parenteral drug product, whereas even trace levels of elemental contamination can severely impact certain processes in semiconductor wafer fabrication facilities. Wipers are used widely in several industries as a part of cleaning protocols that are prescribed in the maintenance of the controlled environment. They are used to clean hard surfaces, equipment, chambers, and tools, clean up spills, or serve as a work surface. Given that wipers serve such integral functions within cleanrooms, there is expectedly a high emphasis placed on their cleanliness. Wipers with lower levels of cleanliness can themselves serve as sources of contamination. Therefore, it is critical that wiper quality and cleanliness levels be adequately measured and comparably evaluated.

4

METHODS TO ASSESS WIPER QUALITY

The quality of a cleanroom wiper is typically evaluated across a range of performance characteristics including fabric substrate and microstructure, sorption

566 Developments in Surface Contamination and Cleaning

capacity and rate, particulate burden of various sizes, bioburden, ions, metals and nonvolatile residues (NVRs), among others. Personnel responsible for the controlled environment typically make an informed judgment about which of these performance attributes are more critical to them compared to others. Test protocols exist to facilitate common methods for the evaluation of wipers from the results of these tests. IEST-RP-CC004.3, “Evaluating Wiping Material Used in Cleanrooms and Other Controlled Environments” [5] describes the different types of contamination related to wiper cleanliness. The three major types of contamination described are particles and fibers, ions, and nonvolatile extractable matter.

4.1 Particles and Fibers IEST-RP-CC004.3, Section 6, describes two methods for particle and fiber enumeration. The overall process for particle and fiber counting is to first extract the particles into solution and then count the extracted particles. The test solution can be pure water or water with an additive to lower the surface tension of the solution and improve the wetting of the polymer surface.

4.1.1 Extraction Some type of motion is used to move particles from the wiper into the extraction solution. The theory is that with more intense motion used in particle extraction, the more particles will be available for enumeration. The less intense motion is used to estimate the number of releasable particles. These particles are on the surface of the wiper or the yarn that comprises the wiper. Different materials of construction have different levels of releasable particles. As a trend, natural fibers such as cotton and wool yield more particles than man-made fibers like polyester. A more intense motion is used to estimate the number of releasable plus generated particles. The wiper undergoes motion in solution, touches the sides of the container, and rubs itself – activities that can generate particles. Typical motion generators are an orbital (less intense) or a biaxial (more intense) shaker as shown in Fig. 14.9. The orbital shaker motion generates small waves that sluice over the wiper. If the motion is too intense, the water will spill out of the photographic tray limiting the energy imparted to the wiper. The biaxial shaker generates notion that is side-to-side and up-and-down. The container is enclosed to prevent the liquid from exiting the container (and to minimize contamination from the environment). When the motion is generated by an orbital shaker, a surfactant can be used, as the extraction solution is filtered before analysis by optical and scanning

Cleanroom Wiper Applications Chapter

14

567

FIGURE 14.9 (a) An example of a wiper undergoing agitation in an orbital shaker. Notice the gentle motion of the water banking across the wiper. (b) An example of a wiper undergoing agitation in a biaxial shaker. Notice the more intense motion of the standing wave at the surface of the water. The wiper is also folding over onto itself.

electron microscopy. A more detailed explanation of this extraction and measuring test method is presented in ASTM Standard E2090, “Standard Test Method for Size-Differentiated Counting of Particles and Fibers Released from Clean Room Wipers Using Optical and Scanning Electron Microscopy” [6]. When the motion is generated by a biaxial shaker, the wiper is extracted using water only. The typical instrument for small particle analysis (0.2 μm or 0.5 μm) is a liquid particle counter (LPC). Because an LPC operates by light scattering, using a surfactant would generate bubbles that will interfere with the particle counting.

4.1.2 Steps of the Extraction Procedure using a Biaxial Shaker and the Liquid Particle Counter This procedure is most common due to the availability of the LPC instrument. The extraction process starts with preparation of the blank where the whole extraction system except for the sample is tested for its particle burden. An acceptable blank is less than ten percent of the sample particle burden. Lower levels are easily achieved. The sample preparation is performed in an ISO Class 5 particle hood to reduce sample contamination. The procedure starts with adding 600 mL of reagent water (deionized, filtered) to a 2-liter Erlenmeyer flask. Using gloved hands and clean forceps, the opening to the flask is covered with aluminum foil that has been rinsed with water to remove loose particles. The flask is mounted onto the biaxial shaker and is secured in place using cushions to prevent the flask from breaking, chipping, or flying out of the shaker. The flask is shaken

568 Developments in Surface Contamination and Cleaning

for five minutes. The flask is removed from the shaker taking care to keep the aluminum lid on the top of the flask. Under the particle hood, the foil lid is removed and a 200-mL sample is poured into a 250-mL clean beaker, reserving the remaining portion for optical fiber analysis. In the liquid particle counter, a clean stir bar is added to the sample beaker and it is placed on the magnetic stirrer. The rate is adjusted such that the water just mixes. The end of the liquid particle counter sample tube is placed into the water and the analyzer. is activated. The cumulative number of counts/mL is recorded at the size of interest (typically >0.5 μm). The counts should be less than 25 particles/mL. If the counts are higher, the glassware should be rewashed and the above steps are repeated. If the result is below 25 particles/mL, a wiper can be analyzed by adding the wiper after 600 mL of water are added to the 2-liter Erlenmeyer flask. The equation to determine the number of particles per square meter of wiper is given below: Particles f½ðSC
df Þ  BC 
V g ¼ m2 Area

(14.1)

where SC is the averaged cumulative sample counts at the size range of interest; df is dilution factor for any external dilution of the sample extract; BC is the averaged cumulative method blank counts in the size range of interest; and V is the total volume of the sample extract. Area ¼ Area of the wiper in m2 An example calculation follows:

Particle range:

 0.5 μm

Particle channels:

0.5–20 μm

Area of wiper:

0.230 m
0.230 m

Average count (Sc):

300 particles/mL

Total volume (V):

600 mL

Blank (Bc):

10 particles/mL

External dilution

No (df ¼ 1)

Particles f½ð300
1Þ  10
600g ¼ 3:3
106 0:230
0:230 m2

Cleanroom Wiper Applications Chapter

14

569

4.1.3 Steps of the Procedure for Using an Orbital Shaker and the Scanning Electron Microscope This procedure is less commonly used by wiper suppliers; however, more sophisticated wiper users employ this method to measure particle release below the orbital shaker energy level. Below is a summary of the ASTM E2090 test method. References to equipment and supplies are better defined in the ASTM test method. The sample preparation consists of four parts. The first part is the preparation of a 0.1% stock solution of a nonionic surfactant. TRITON® X-100, an octylphenol 9.5 mole ethoxylate, is frequently used; however, any ethoxylated alkyl phenol can be used. The choice of extraction fluid is not restricted to only using a surfactant. DI water or another substance to reduce the surface tension of water, e.g., six percent isopropyl alcohol, can be used. In an ISO Class 5 particle hood, one gram of TRITON® X-100 is added to 1000 mL of particle-free distilled water in a clean beaker. A magnetic stir bar is added to the beaker which is covered with a pre-rinsed aluminum foil. The beaker is gently warmed to 313–323 K (40–50 °C) while stirring until the surfactant is in solution. After the surfactant is in solution, the beaker is removed from the hot plate and allowed to cool in the particle-free hood while covered. If the solution is heated above the cloud point of the solution, the glassware must be cleaned and the solution is remade. The second part is preparation of the background filter stub. This step tests the extraction system for its particle cleanliness. All the glassware and the plastic photographic tray are thoroughly cleaned. The tray is placed on the platform of the orbital shaker and 500 mL of deionized water are added. A 25-mL aliquot of the 0.1% surfactant solution is added to this water. The tray and water are agitated for five minutes. While the tray is in motion, a 25-mm diameter polycarbonate filter is mounted onto the face of the stainless steel screen of a vacuum filtration apparatus. The assembly of the filtration system is completed and the solution is poured through the filtration system. It is important that the water level is kept high in the funnel of the filtration system during the addition of the solution. This aids in keeping the particle level consistent across the surface of the filter. After all the solution is filtered, the filtration apparatus is disassembled, the filter is removed using clean tweezers, and is placed on a SEM stub to dry in the ISO Class 5 particle-free hood. Once the filter is dry, colloidal graphite in isopropyl alcohol solution is applied in spots around the filter edges to aid in attaching the filter to the SEM stub and in electron conduction to ground from the filter. When the stub is dry, the stub surface is again sputter coated with a metal, e.g., gold, to aid in the conduction of electrons from the filter surface. The next part is evaluating the SEM stub for its particle level. Once the filter stub is coated, the entire stub surface is examined using an optical microscope. A typical stereomicroscope is shown in Fig. 14.10. A calibrated reticule is used

570 Developments in Surface Contamination and Cleaning

FIGURE 14.10 A stereomicroscope with a light source at a glancing angle.

for sizing large particles and fibers. The purpose of this step is to confirm that the fibers and large particles are reasonably evenly distributed across the surface of the filter and that there are only few particles and fibers. There should not be a deposit of particles and fibers around the edge where the funnel meets the filter. If the fibers and particles are not distributed evenly or there are many fibers, the stub is discarded and the above steps are repeated. How the light strikes the surface of the stub is important for evaluating the surface. For the best view of fibers and large particles, the light should be set a low, glancing angle to the surface of the stub as shown in Fig. 14.10. This angle increases the contrast of the darker metal coated filter and the almost clear plastic fibers for improved viewing through the reduction of the glare from reflected light from the surface of the stub. If the distribution is even and there are only a few fibers, the surface is scanned and the fibers on the surface are counted. A typical count is less than six fibers. If the background stub meets the criteria, the stub is loaded into the SEM for further evaluation for the number of particles in the smaller size ranges.

4.1.4 Enumerating Small Particles Two methods of particle enumeration are described in the recommend practice (RP): liquid particle counting (Fig. 14.11), a light scattering process, and

Cleanroom Wiper Applications Chapter

14

571

FIGURE 14.11 An example of a SEM in use.

FIGURE 14.12 An example of a liquid particle counter in use.

scanning electron microscopy (Fig. 14.12). Once the particles are in solution, the number of particles is assessed in the form of particle counts. A review of the theory behind particle counting by LPC and by SEM follows. Liquid Particle Counting The theory behind liquid particle counting is the scattering of light by small particles. The original theory [7] was introduced by Gustav Mie as a solution to Maxwell’s electromagnetic equations for light scattering off spherical

572 Developments in Surface Contamination and Cleaning

particles. The solution is mathematically complicated. In summary [8], the intensity of the light scattering off a spherical particle is a function of the index of refraction for the medium and for the particle, the diameter of the particle, the angle of detection, and whether the particle absorbs at the incident wavelength. Liquid particle counters are typically calibrated using polystyrene spheres in water. The identification of the polystyrene particle in water sets the index of refraction for the particle and the medium, respectively. This calibration and the instrument construction (laser wavelength, light collection geometry, light detection, electronics) also fixes the amount of light collected and turned into an electrical signal. The article and book by Particle Measuring Systems reviews the particle light scattering theory, gives examples, and shows applications [9,10]. The amount of signal is processed by electronics and related to a particle size. Software allows the binning of particle size ranges for calculations and reports. As the sample is introduced through the particle analyzer, light scatters off objects, e.g., undefined particles and air bubbles, which have an index of refraction different from the continuous medium (water). If the particles absorb at the laser wavelength, the amount of light scattered is reduced due to absorption, appearing as a smaller particle. Mie’s theory was developed for spherical particles. Particles extracted from wipers are rarely spherical. The amount of light scattered is not related to the smallest or largest dimension, but somewhere in between. The number of particles passing through the evaluation volume must be limited to one. Two particles scatter more light than one which gives a false reading of one larger particle instead of two smaller particles. Sample dilution reduces the risk of this event from occurring.

Scanning Electron Microscopy The principles behind SEM are different from the light scattering principles used in liquid particle counting [11–14]. In the SEM, electrons are emitted as a spray from an electron gun. The energy of the electron beam is determined by the potential difference between the electron source (the cathode) and the anode. The accelerated electron beam travels through electromagnetic lenses and deflector coils that focus and move the beam in a raster pattern across the sample surface. A diagram of the inner working of an SEM is available on the Internet as shown in [15]. Different signals are produced from the electron beam impinging on the sample surface. The list of these signals includes, but is not limited to, characteristic x-rays, backscattered electrons, and secondary electrons. The secondary electron signal is generated by the ejection of low energy (compared to the electron beam energy) electrons from surface or near-surface atoms. These free electrons are collected using a positive voltage screen to

Cleanroom Wiper Applications Chapter

14

573

accelerate the electrons towards the detector. In the detector, the electron signal is converted to an electronic signal. The image is formed through the electron beam which moves in a raster pattern interacting with the surface electrons of the sample. Higher atomic number atoms are more likely to eject an electron, therefore making more electrons that are available for image formation. The type of atoms and the surface features together modulate the number of electrons that are free to be detected and converted into an electronic signal. The electronic signal is the intensity portion of the image viewed on the SEM screen. The raster pattern of the electron beam matches the raster pattern of the television monitor. Increased magnification is achieved by reducing the path length of the raster pattern of the incident electron beam. As the path length of the raster pattern shortens, the magnification of the sample surface increases.

4.1.5 Enumeration of Particles and Fibers Fibers are enumerated using an optical microscope at 40
magnification. As the sample is condensed to a one-inch diameter filter, the whole surface of the filter is evaluated for the number of fibers. The equation to determine the number of fibers per square meter for a wiper is given below. Fibers SC  BC (14.2) ¼ Area m2 where SC are the sample counts at the size range of interest; BC are the blank counts at the size range of interest; Area is the area of the wiper in m2. An example calculation follows: Area of wiper (m2):

0.230 m
0.230 m ¼ 0.0529 m2

Number of fibers counted (wiper):

90

Number of fibers counted (background):

4

90  4 1630 fibers ¼ 0:0529 m2 m2 The SEM is used to enumerate two size ranges of particles. The small particles range in size from 0.5 μm to 5 μm. This is achieved at 3000
magnification. The large particles range in size from 5 μm to 100 μm. This is achieved at 200
magnification. The general equation used to determine the number of particles per square meter of wiper for the SEM is given below.

574 Developments in Surface Contamination and Cleaning

    PS PB 
NF Particles FS FB ¼ Area m2

(14.3)

where FB is the number of background fields counted; PB is the total number of background particles counted; FS is the number of sample fields counted; PS is the total number of sample particles counted; NF is the total number of fields; and Area is the area of the wiper. The number of fields is determined by the ratio of the area of the filter and the field of view of the SEM at a given magnification. Sample calculations for determining the number of small and large particles using the SEM are given below. A. 3000
Magnification: Particles 0.5–5.0 μm     PS PB 
NF Particles 3000X FS FB ¼ Area m2 Number of background fields counted (FB):

32

Total number of background particles counted (PB):

10

Number of sample fields counted (FS):

16

Total number of sample particles counted (PS):

75

Total number of fields (NF):

155 251 0.23 m
0.23 m ¼ 0.0529 m2

2

Area of wiper (m ):

 Particles 3000X ¼ m2

   75 10 
155; 251 Particles 16 32 ¼ 12:8 x 106 0:0529 m2

B. 200
Magnification: Particles 5.0–100 μm Number of fields counted (background):

16

Total number of particles counted (background):

10

Number of fields counted (wiper):

8

Total number of particles counted (wiper):

87

Total number of fields of view at 200


688

Area of wiper (m2):

0.23 m
0.23 m ¼ 0.0529 m2

Cleanroom Wiper Applications Chapter

 Particles 200X ¼ m2

14

575

   87 10 
688 Particles 8 16 ¼ 133 000 m2 0:0529

4.1.6 Other Considerations So far, the discussion has centered on the contamination level of the particles and fibers found on the wiper. While understanding the contamination level of the wiper is important, what needs to be understood is what level of contamination the wiper leaves behind on the work surface. There are many variables that must be managed to begin evaluating this transfer, such as pressure, speed of wiping, surface roughness, wiper construction, etc. While the wiper leaves some contamination behind, there is nothing like using a wiper to clean a surface. The surface contamination is transferred from the surface to the wiper in a controlled manner, it is not being scattered, and is able to be removed from the cleanroom. 4.2

Ions

Extractable or leachable ions can remain as contaminants after the fabric is processed. Typical cations are sodium, potassium, calcium, magnesium, and ammonium. Typical anions are bromide, chloride, fluoride, nitrate, nitrite, sulfate, and phosphate. These ions can be extracted using pure water. As the manufacturing environment for microelectronics products and the sensitivity of allergic metals, e.g., chromium, nickel, silver and copper, causing health issues, such as contact dermatitis, hypersensitivity, eczema, etc. [16–20] in parenterals and medical devices becomes more stringent and more intolerant of metal contamination, understanding the trace metals burden of wipers is required. It is important to accurately estimate the ionic burden that the wiper carries with it into the cleanroom. For trace metals analysis, hydrochloric, nitric and hydrofluoric acids may be required to bring these metals into solution for analysis. The ionic levels are typically determined by ion chromatography (IC) using conductometric detection. Other analytical techniques may include the use of inductive coupled plasma optical emission spectrometry (ICP-OES), inductive coupled plasma mass spectrometry (ICP-MS), inductive coupled plasma quadrupole mass spectrometry (ICP-QMS) and inductively coupled plasma sector field mass spectrometry (ICP-SFMS). To extract the ions, a wiper is soaked in water at a given temperature for a specified time. The extraction temperature and time are varied to derive different information about the wiper. The extraction temperature can be ambient or elevated (353 K is common). The extraction time, which ranges from 15 minutes to 24 hours, varies with the temperature. Extractions performed at elevated temperature and for a short time estimate the maximum ion contamination that the wiper contains. Extractions performed at ambient temperature are targeted to estimate the amount of ions that may be extracted during use.

576 Developments in Surface Contamination and Cleaning

4.3 Extractable Matter Trace amounts of finishing oils and other additives used in the fabric manufacturing process can remain as extractable or leachable contaminants after the fabric is processed. A wiper is soaked in a solvent at a given temperature for a specified time. The amount of extractable matter for a given solvent is dependent on the time and temperature of the extraction. Typically, the solvent is chosen because the wiper will be used with the solvent. Extracting the wiper at or near the boiling point of the solvent will remove more material. Extracting with the solvent at room temperature will give an estimate of how much material the wiper will leave behind. The wiper is extracted with an excess of solvent at a given time and temperature. The solution is dried and the amount of extracted material is determined gravimetrically. The results are reported as a percent, or by multiplication of the basis weight of the wiper, in grams per square meter.

4.4 Using Box and Whisker Charts to Evaluate Wiper Consistency as the Measure of Quality There are expectations for the products that are purchased for the cleanroom that are often measured in terms of performance satisfaction. This performance is often assessed in terms of quality and consistency of the product. But what do these words really mean? Quality is commonly defined as meeting or exceeding stated performance targets, while consistency is “the agreement or harmony of parts or features to one another or a whole” [21]. Therefore, consistency is an integral parameter that allows the expected quality to be achieved.

4.4.1 Determining Cleanroom Wiper Consistency Wipers are used to control contamination in cleanroom environments in a variety of industries, from building the next-generation microchip to manufacturing the newest vaccine. Each of these settings may have different applications for a cleanroom wiper, but measuring the wiper quality should always be the same, i.e., consistent. Three methods of evaluating wiper consistency using a common sample data set are demonstrated below. Statistical Process Control Statistical process control (SPC) is the application of statistical methods to the monitoring and control of a manufacturing process to ensure that it operates at its full potential to produce a conforming product. Wiper manufacturers should employ SPC programs to control the physical, chemical and contamination characteristics for each wiper lot that is manufactured.

Cleanroom Wiper Applications Chapter

14

577

70

Measurement result

60 50 40 30 20 10 0 0

5

10

15 Sample number

20

25

30

FIGURE 14.13 SPC chart resulting from the evaluation of one product multiple times. Note: The values along the y-axis represent a relative test result. If the test were measuring the particle contamination level of a wiper (IEST-RPCC004.3, Section 6, biaxial shake, >0.5 μm LPC), the y-axis units would be in millions of particles per square meter.

70

Measurement result

60 50 40

Wiper 1

30

Wiper 2 Wiper 3

20

Wiper 4 10 0 0

5

10

15 20 Sample number

25

30

FIGURE 14.14 SPC chart resulting from the evaluation of four products multiple times. Note: The values along the y-axis represent a relative test result. If the test were measuring the particle contamination level of a wiper (IEST-RPCC004.3, Section 6, biaxial shake, >0.5 μm LPC), the y-axis units would be in millions of particles per square meter.

Typically, SPC data are plotted by sample number (as shown in Fig. 14.13). However, if multiple lots or wipers are to be compared, determining the best quality wiper can quickly become confusing and uninformative (as shown in Fig. 14.14). The downside is that with these data sets determining which wiper has the highest quality is often difficult.

578 Developments in Surface Contamination and Cleaning

TABLE 14.3 The Four Wiper Data Sets Represented in Fig. 14.14 are Reduced to Averages And Standard Deviations Wiper 1

Wiper 2

Wiper 3

Wiper 4

Average

12.17

23.28

16.38

22.41

Std. Dev.

3.71

8.62

15.18

16.73

Data Averages and Standard Deviations A commonly used method to compare cleanroom wiper quality is through data averages and standard deviations which assumes a normal distribution. (Note: The same data set as represented in Fig. 14.14 is compiled in Table 14.3 as averages and standard deviations.) Such a method reduces a very large set of available data that has been produced over time to two numbers that inadequately represent the data set. Each data set is summarized by just two data points, and, as a result, much information is lost.

Box and Whisker Chart Incomplete data summaries as represented by typical or average values misrepresent the true quality of a cleanroom wiper. A quicker, easier, and more statistically unbiased method to evaluate many large sets of data is a Box and Whisker Chart [22]. These charts represent data sets pictorially. The components of a Box and Whisker Chart are: l

l

l

l

Line – represents the median or middle value of a ranked data set. (Extreme values do not affect the median value as much as an average could be affected.) Box – represents the range of values within which fifty percent of the data lie. If the median line is nearer to one end of the box, the data are skewed toward that end. A smaller box indicates that the values are more similar. Whisker – the line at each end of the box, expresses a range of values in which twenty-five percent of the data lie. A short whisker indicates that values within the whisker range are similar to each other. Outlier – indicates values that are significantly different than the rest of the data set.

These charts are constructed through the following steps. 1. The data set values are ranked from highest to lowest. 2. The ranked data are divided into quartiles.

Cleanroom Wiper Applications Chapter

14

579

a. The median value (Q2) is the middle value of the entire data set, which, in general, is the most likely value in the data set. b. The first quartile value (Q1) is determined by locating the median of the lower half of the data set. c. The third quartile value (Q3) is found by determining the median value of the upper half of the data set. 3. The box is constructed. a. Use the Q1 and Q3 values as the lower and upper bounds of the box. 4. The whisker ends are defined. a. The difference of the box ends, Q3 – Q1, defines the Intra-Quartile Range (IQR). b. The lower whisker is determined by Q1 – 1.5*IQR. If the last value of the data is greater than the value determined by Q1 – 1.5*IQR, the whisker is shortened to that value. c. The upper whisker is determined by Q3 + 1.5*IQR. If the last value of the data is less than the value determined by Q3 + 1.5*IQR, the whisker is shortened to that value. 5. The outliers are determined. a. Any values beyond the whiskers are considered outliers (values that are markedly smaller or larger than other values) and are indicated with an asterisk. Fig. 14.15 shows a simple mock data set. Shown below are calculations that determine the median value, the values of the box ends, the values for the upper and lower whiskers, and the indication of an outlier data point. 40 35 Whisker

Q3

30 Median Count

25 Q2 20 15 Q1 Box

10 5

Outlier

35 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 1

0 FIGURE 14.15 Box and whisker chart with its components labeled for a mock data set shown on the right.

580 Developments in Surface Contamination and Cleaning

l l l l

l

l

The median or middle value is 19. The box end values are 15 and 23, Q1 and Q3, respectively. IQR ¼ Q3 – Q1 ¼ 23 – 15 ¼ 8 Lower whisker ¼ Q1 – 1.5*IQR ¼ 15 – 1.5*8 ¼ 15 – 12 ¼ 3. The lower whisker is located below the value “12” at “3.” Upper whisker ¼ Q3 + 1.5*IQR ¼ 23 + 1.5*8 ¼ 23 + 12 ¼ 35. The upper whisker is located at the value “35.” As the value “1” is beyond the lower whisker, it is marked with an asterisk (*) as an outlier.

Interpreting the Box and Whisker Charts Fig. 14.16 shows a comparison of the data sets for four wipers. The data were obtained by the method described in IEST-RP-CC004.3, Section 6, biaxial shaker, >0.5 μm LPC analysis of wipers. The particle contamination level as measured by LPC is used here as a demonstration of the power of using a box and whisker chart. However, any wiper property can be charted to understand its consistency. Looking at the chart (Fig. 14.16), the following observations can be made: l l l l

Wiper 1 has the smallest box and the shortest whisker. Wiper 2 and Wiper 4 have similar medians. Wiper 3 has the lowest median. Wiper 3 has an outlier value as shown by the asterisk above the whisker and the longest whisker.

70.0

Biaxial shake, >0.5 µm LPC, wiper comparison

60.0

Particles/m2, ×106

50.0

40.0

30.0

20.0

10.0

0.0

Wiper 1

Wiper 2

Wiper 3

Wiper 4

FIGURE 14.16 Box and whisker chart comparing IEST-RPCC004.3, Section 6, biaxial shake, >0.5 μm LPC analysis data sets for four wipers.

Cleanroom Wiper Applications Chapter

l

l

14

581

The median for Wiper 3 is nearer to the lower end of the box indicating many values are similar and low, that is, fifty percent of the data values are between 2 and 8; however, the other fifty percent of the values range between 8 and 58. Wiper 4 has the largest box and the largest range in the data.

Summarizing these observations, Wiper 1 and Wiper 3 are the better products of the four. In comparing Wiper 1 and Wiper 3, the following observations can be made: l

l l

The data set for Wiper 1, or the whole box and whisker diagram, lies within the box for Wiper 3. Of the test results for Wiper 3, 25% are lower than those for Wiper 1. More than 25% of the rest results for Wiper 3 are higher than those for Wiper 1.

Box and whisker charts present a far better means to fairly evaluate the quality of cleanroom wipers than simply comparing the average value and/or the standard deviation which are necessarily incomplete representations of a data set. A minimum of three data points is necessary to define the construction of a box and whisker chart. However, when evaluating cleanroom wipers in practice, the box and whisker charts being compared should have been constructed from more than 15 data points each, as well as comparable numbers of data points for each for the analysis to be most meaningful. The true measure of the quality of a cleanroom wiper lies in its consistency as manufactured over an extended period (i.e., months, years) using a given, unaltered process. What truly matters in a critical cleaning operation is that each wiper from a bag, each bag within in a lot, and each lot of a given wiper product is delivered to the end user with the highest assurance of the expected quality. Box and Whisker charts offer the most unbiased representation of the consistency of cleanroom wipers from within a bag or lot, over an extended period. In conclusion, selecting the best cleanroom wiper for a particular application requires the most unbiased scientific assessment of the available data for any given wiper. The Box and Whisker chart comparison of the four wipers shown here allows for a quick determination that Wiper 1 is a better performing cleanroom wiper because it is more consistent in its quality measures. A user has greater assurance that Wiper 1 will perform as expected with a higher degree of confidence when compared to the other wipers shown in the data set due to its greater consistency. The quality of a cleanroom wiper should, therefore, be evaluated not merely through a typical or average value, but more importantly through a statistically valid assessment of how consistently that typical value is attained in practice over a period of time.

582 Developments in Surface Contamination and Cleaning

4.5 Advantages and Disadvantages of Wiper Test Methods A comparison of the advantages and disadvantages of the use of LPC or SEM for particle sizing and enumeration is shown below (Table 14.4). After the particles are extracted from the wiper, the extract solution can be analyzed in less than 5 minutes. The SEM sample preparation after particle extraction requires time for the filter to dry on the SEM stub and gold sputter coating. After these preparation steps, the sample chamber must be evacuated, requiring even more time before enumeration can begin. Fortunately, some SEMs allow multiple samples to be introduced for enumeration, thus saving evacuation time for the other samples. Routine analysis, which does not include maintenance or troubleshooting, is simpler with the use of the LPC instrument. Routine analysis for SEM requires more skill and intuition to adjust the image clarity for accurate particle enumeration. For the two size ranges described in the calculations above, interpretation of the particle images for their size requires training and sizing aids. Some technician interpretation affects the enumeration result. After enumerating several samples, eye strain can become an issue.

TABLE 14.4 A Comparison of the Advantages and Disadvantages of the Use of a LPC or SEM Mode

LPC

Advantages

l

l

l

l

Disadvantages

l l

l

Commonly used – many companies have a large database as reference Lower priced, allows more users for wiper quality assessment Quick sample preparation and analysis is faster Easy to use Particle identity is hidden Particle may be mis-sized due to absorption of light, or having a different index of refraction Possible entrained air may increase particle counts

SEM l l

l

l

l

l

l

l

Most direct view of particles Increased magnification allows evaluation of smaller particles Larger particles can be enumerated With added instrumentation, elemental analysis of a single particle can be performed Rarely used for particle counting Higher price, a barrier for common usage Sample preparation and analysis longer Analyst requires more training to use. Many adjustments are available to improve image quality

Cleanroom Wiper Applications Chapter

5

14

583

THE IMPORTANCE OF AUTOMATION

One of the key determinants of the cleanliness levels of cleanroom wipers is the level of automation used in their manufacture. Over the past decade, cleanroom wiper manufacturing technology has made great strides in reducing the contamination burden carried by these wipers. A major contributor to these improvements in cleanliness has been the advances made in automated and semiautomated manufacturing of cleanroom wipers. Beginning with the variety of edge treatments used, to the fully automated manufacturing of cleanroom wipers, automation is a key contributor to the cleanliness of the resulting wiper being used. Given that manual processing and personnel are often the primary source of contamination in cleanrooms, it is logical to recognize the importance of the automated manufacture of cleanroom wipers.

5.1

Wiper Edge Treatments

Applying different edge treatments to wipers can have differential impacts on the cleanliness of the resulting wipers. As shown in Fig. 14.17, particle and fiber burden of a wiper can be significantly impacted by the nature of the edge treatment. Cut edge polyesters will typically be higher in both small particle and large fiber burdens compared to thermally cut wipers, which in turn will be higher than sealed-border wipers. Fig. 14.17 shows the relationship between the fiber and small particle generation when wipers are formed by different manufacturing methods. The “Cut Edge” manufacturing process generates small particles while fibers are shed from the rough edge. The “Hot Cut” manufacturing process generates small particles through the knife crushing the yarn in the fabric, but the thermal energy melts the ends of the fabric yarns. The “Sealed Border, Laser Edge” manufacturing process melts the polymer yarn producing few small particles and seals the edges of the wiper to create the sealed border.

5.2

Automated Manufacture of Wipers

It is not commonly recognized that cleanroom wipers have traditionally been manufactured manually by hand. This “conventional laundry” operation is often carried out by hand with many people, gowned and gloved, inspecting and stacking wipers, and then packaging them into the bags. It is somewhat counterintuitive to require ultraclean wipers going into sensitive cleanroom environments, but to then recognize that they are usually contaminated with the residual burden of being handled by people during manufacture. As a result, several efforts have been made over the years to achieve lower levels of particulates and higher levels of consistency using automation instead of people. Shown in Fig. 14.18 is the clear benefit of increased median levels of cleanliness and greater consistency achieved using automation.

584 Developments in Surface Contamination and Cleaning

>100 mm fibers versus >0.5 mm LPC edge treatment 4,500

>100 mm optical microscopy, fibers/m2

4,000

3,500

Cut edge 3,000

2,500

2,000

Hot cut 1,500

1,000

Sealed border, laser edge

500

0 0.00

5.00

10.00

15.00

20.00

25.00

30.00

>0.5 mm LPC, ¥10 particles/m 6

2

FIGURE 14.17 This scatter chart compares the IEST-RP-CC004.3, Section 6, >100 μm optical microscopy analysis test results to the IEST-RP-CC004.3, Section 6, biaxial shake, >0.5 μm LPC analysis test results for wipers manufactured by different edge treatments. A “Cut Edge” wiper is formed using a sharp knife to cut the fabric. A “Hot Cut” wiper is made by cutting the wiper with a thermally heated rotary knife. A “Sealed Border, Laser Edge” wiper is formed by melting a small width of the wiper edge either by ultrasonics or by laser.

Fig. 14.18 shows a comparison of data sets for two wipers. The data were obtained by the method described in IEST-RPCC004.3, Section 6, biaxial shake, >0.5 μm LPC analysis of wipers. The wiper identified as “Automated” is manufactured using a process where humans have intermittent contact with the product. The product identified as “Conventional laundry” is made in a conventional cleanroom laundry where each wiper is exposed to humans and the environment. Even a partially automated process reduces the variability in a product.

Cleanroom Wiper Applications Chapter

14

585

>0.5 µm LPC data showing automation effect 20

>0.5 µm LPC, ×106 particles/m2

18 16 14 12 10 8 6 4 2 0 Automated

Conventional laundry

FIGURE 14.18 A consistency chart, which compares the IEST-RPCC004.3, Section 6, biaxial shake, >0.5 μm LPC analysis test results for two wipers, is shown above. The same fabric, both 100% polyester with the same knit structure, was manufactured through different cleaning processes. The product labeled “Automated” has intermittent human contact. The product labeled “Conventional laundry” is processed through a typical cleanroom laundry.

6

APPLICATIONS

The selection of wipers for use in cleanroom applications is critically dependent on the specific nature of the residues that need to be removed and the nature of the surfaces from which they must be removed. In some instances, wipers may be used to apply cleaning agents or disinfectants to surfaces. In general, this translates into certain overarching needs for the properties of the cleanroom wiper used across a variety of industries. Below we highlight some of the specific and unique requirements in a variety of applications that may drive wiper selection and use in different industries.

6.1

Semiconductors

Product yield losses in semiconductor fabrication facilities (fabs) are often driven by contamination that enters the fab environment through external sources. This makes it critical for fab operations to ensure that all sources of contamination external to the environment including supplies are adequately scrutinized for intrinsic cleanliness, i.e., the contamination burden may contribute to the environment during use. The requirements for wipers range from ultra-clean premium wipers that carry extremely low background metal burden to prewet wipers to be used in gowning ante-rooms that may have higher

586 Developments in Surface Contamination and Cleaning

intrinsic levels of particulate contamination, but are suited for use in environments that have higher tolerance for such contamination. Wipers used for cleaning fabrication chambers and the variety of tools used in chemical processes involved in the silicon chip manufacture, must have extremely low levels of particulate and fiber burden. In addition, they must also have extremely low levels of ionic burden and NVRs. In use, these wipers will typically be pre-wetted with cleaning solutions such as IPA (isopropyl alcohol) to effectively remove contamination from surfaces. To maintain consistent product yields, it is critical that these wipers do not introduce any extraneous contamination into the fab as such contamination can migrate into the mini-environments contacting the final product. In general, it is considered too risky to introduce any wipers other than knit polyesters into the environment.

6.2 Disk Drives Data storage media manufacturing environments are particularly sensitive to ionic and particulate contamination. It is critical that wipers and swabs used in these areas and around these products do not contribute extraneous contamination burden. The use of knit polyesters is recommended wherever possible. Ultraclean polyester swabs of varying head shapes are particularly useful in reaching into narrow areas and in cleaning specific small geometries.

6.3 Pharmaceuticals The process of manufacture of active pharmaceutical ingredients (APIs) (drug substances) is a critical step in the manufacture of the final drug product. The intervening steps often involve mixing, blending, formulation, tableting, and fill/finish. The cleanliness requirements in each of these areas may be different, often depending on the chemical nature of the materials being processed. Typically, the cleanliness requirements of wipers used in these areas will be more in line with ISO Class 5-7 characteristics described in Table 14.1 above. Wipers used may be either dry or prewet. The proper application of qualified disinfectants using wipers and cleaning validation using cleanroom swabs are an FDA requirement in these areas according to cGMP (current Good Manufacturing Practices).

6.4 Biologics The manufacture of biologics is a fast-growing area with increased focus on contamination control. Of primary importance in this area is the microbial contamination. Given that the production of biologics is often derived from bacterial or fungal cultures in aqueous media, these environments and the resulting product may be especially prone to microbial contamination. Most processing of biologics is conducted in aseptic suites and many of the processes that are

Cleanroom Wiper Applications Chapter

14

587

followed are aseptic processes. All the supplies that are used to manufacture biologics in an aseptic suite are typically presterilized by gamma irradiation, ETO (ethylene oxide) or autoclaving prior to use. Likewise, cleanroom wipers used in sterile suites must be presterilized using gamma irradiation to minimize the risk of bioburden being introduced into the cleanroom. In addition, it is critical to the safety of the biologics drugs manufactured that the wipers not introduce any external source of endotoxins to the aseptic suite. As a result, these sterile cleanroom wipers must often be tested for low endotoxin burden of under 20EU per AAMI (Association for the Advancement of Medical Instrumentation) guidelines. Wipers that are gamma irradiated to be sterilized must undergo sterile validation to ensure that the irradiation has the desired effect and ensures a Sterility Assurance Level (SAL) of 10-6. Typical applications for sterile cleanroom wipers are the wiping of process and cleaning residues using dry or prewet wipers, and the application of disinfectants. Sterile microdenier wipers are particularly effective at removing contamination including the residues from biofilms. In general, it is advisable to use sterilized polyester wipers, given the highly fragile nature of the products being manufactured, but there may be areas in which a nonwoven substrate may be appropriate.

6.5

Medical Devices

The manufacture of medical devices typically uses a terminal sterilization procedure to minimize the risk of microbial contamination. As such, the production environment itself may not be aseptic in nature, but it is nevertheless susceptible to physical and chemical contamination from a variety of external sources. Therefore, it is critical that wipers and swabs used to clean surfaces and components of medical devices during manufacture are selected to ensure that they do not leave behind particulates and other residues that can contaminate the end product. Ultraclean swabs are particularly critical for use in medical devices since it is often necessary to access narrow crevices and channels. Similarly, the larger work surfaces in a medical device manufacturing environment also need to be wiped and mopped to ensure that contamination does not migrate towards the product.

7

CURRENT TRENDS IN WIPER TECHNOLOGY

Wiper manufacturers are no different than other industries, trying to optimize current processes to provide a cleaner product while containing costs in light of ever-increasing costs of raw material components. Wiper manufacturing and performance characteristics are evolving as the industries that use wipers evolve. The microelectronics and aeospace industries are demanding overall cleaner products, as the process needs have less room for error and the cost of yield or product loss continues to increase. To meet the ever-changing needs, particle counts are moving to smaller sizes. 0.5 μm has been considered the

588 Developments in Surface Contamination and Cleaning

standard size of the smallest particle counted historically, but many users and manufactureres are now sizing and counting particles as small as 0.1 μm. Currently, testing is being done by LPC. Along with testing for smaller particle sizes, the industry, from both a user and manufacturer standpoint, is working to standarize testing and reporting protocol in conjunction with the Institute of Environmental Sciences and Technology (IEST). Currently, there are many different test methodologies and reporting formats for what seems to be the same physical attribute. Because of the different procedures and reporting formats, wiper users may not be able to easily compare products based on data provided by wiper manufacturers.

8 FUTURE DEVELOPMENTS IN CLEANROOM WIPERS The semiconductor industry has become even more aware of possible contamination impact as the geometry of semiconductor chips continues to shrink. Chip manufacturers are becoming concerned with not only contamination characteristics of a wiper, but the contamination profile of the particle. This may include trying to reduce or control the metal content of synthetic wipers. The difficulty with this is that synthetic wipers are derived from oil and natural gas. The metal content of each gas or oil field is different which means the metal content will fluctuate over time as input supplies to make polymers change. There is a way to clean polyester fiber to very low detectable levels (<10 parts per billion), but this has not been successfully converted to production scale in an economical manner. Wiper manufacturers are also working on products to effectively contain and remove nanoparticles. Nanotechnology and materials are relatively new and, as such, recommended cleaning protocols are just being developed, understood, and implemented. Traditional materials used for wiper construction may or may not work as effectively in capturing particles in the sub-0.5 μm range, so new materials are being developed to better capture nanoparticles. One such option may be the transfer of technology from air filtration to wiper substrates. For years, the life science industry has been concerned with bacteria in critical environments and the possible negative impact on the end product, whether it is a tablet, injectable drug, or implant device. Other industries, such as aerospace and microelectronics, are also becoming aware of the dangers of bacteria, not necessarily in the traditional sense of microbial contamination, but as particulate contamination. This is especially true for equipment bound for outer space, where degrading bacteria may interfere with the inner workings of sensors or other critical parts. As a guard against this, many users in the life science markets have converted to the use of sterile wipers. The particle may still be on the wiper, but it is no longer living to present a threat. Some of the most critical applications may consider the use of sterile products where they traditionally have not been used. At the same time, wiper manufacturers are going to great

Cleanroom Wiper Applications Chapter

14

589

lengths to minimize the potential threat of bacteria and microbes on non-sterile product. This is being done by controls in the manufacturing environment and by minimizing or even eliminating the human interface to manufacture a wiper.

REFERENCES [1] R. A. Bowling, “A Theoretical Review of Particle Adhesion”, in: Particles on Surfaces 1: Detection, Adhesion, and Removal, K. L. Mittal (Ed.), pp. 129–142, Plenum Press, New York, NY (1988). [2] Y. Igarashi and H. Ito, “Double-Knit Fabric Having Non-Run and Stretchability Characteristics and Method and Apparatus for Knitting the Same”, U.S. Patent 5,463,881 (1995). [3] M. W. Wren, M. S. Rollins, A. Jeanes, T. J. Hall, R. G. Coe¨n and V. A. Gant, “Removing Bacteria from Hospital Surfaces: A Laboratory Comparison of Ultramicrofibre and Standard Cloths”, J. Hosp. Infect. 70, 265 (2008). [4] ISO 14644-1, “Cleanrooms and Associated Controlled Environments—Part 1: Classification of Air Cleanliness by Particle Concentration”, International Organization for Standardization, Geneva, Switzerland (2015). [5] IEST-RP-CC004.3, “Evaluating Wiping Materials used in Cleanroom and Other Controlled Environments”, Institute of Environmental Sciences and Technology, Schaumburg, IL (2004). [6] ASTM E 2090-12, “Standard Test Method for Size-Differentiated Counting of Particles and Fibers Released from Cleanroom Wipers using Optical and Scanning Electron Microscopy”, ASTM International, West Conshohocken, PA (2012). [7] G. Mie, “Beitr€age zur Optik tr€uber Medien, speziell kolloidaler Metall€osungen”, Annalen der Physik 330, 377 (1908). [8] D. W. Hahn, “Light Scattering Theory”, Department of Mechanical and Aerospace Engineering, University of Florida, Gainesville, FL (2009). http://plaza.ufl.edu/dwhahn/Rayleigh% 20and%20Mie%20Light%20Scattering.pdf. [9] Basic Guide to Particle Counters and Particle Counting, Particle Measuring Systems, Boulder, Colorado (2011). http://www.pmeasuring.com/wrap/filesApp/BasicGuide/file_1/ ver_1317144880/basicguide.pdf. [10] E. Terrell, J. Gromala and D. Beal, “Understanding Liquid Particle Counters - A Comprehensive Review of Optical Particle Counting Technology and Performance Verification as Applied to the Counting and Monitoring of Particles in Liquid Chemicals and Ultrapure Water”, White Paper, Particle Measuring Systems, Boulder, CO (2006). [11] J. Goldstein, D. E. Newbury, D. C. Joy, C. E. Lyman, P. Echlin, E. Lifshin, L. Sawyer and J. R. Michael, Scanning Electron Microscopy and X-Ray Microanalysis, 3rd Edition, Springer Science and Business Media, Inc., New York, NY (2012) [12] P. J. Goodhew, J. Humphreys and R. Beanland, Electron Microscopy and Analysis, 3rd Edition, Taylor & Francis, New York, NY (2001). [13] M. H. Loretto, Electron Beam Analysis of Materials, 2nd Edition, Chapman & Hall, New York, NY (1994). [14] L. Reimer, Scanning Electron Microscopy: Physics of Image Formation and Microanalysis, 2nd Edition, Springer-Verlag, Heidelberg and Berlin, Germany (1998). [15] Scanning Electron Microscope, Purdue University, West Lafayette, IN (2012). http://www. purdue.edu/rem/rs/sem.htm [16] K. Heim and B. A. McKean, “Children’s Clothing Fasteners as a Potential Source of Exposure to Releasable Nickel Ions”, Contact Dermatitis 60, 1005 (2009).

590 Developments in Surface Contamination and Cleaning [17] J. P. Thyssen, J. D. Johansen, T. Menne, C. Liden, M. Bruze and I. R. White, “Hypersensitivity Reactions from Metallic Implants: A Future Challenge that Needs to be Addressed”, Brit. J. Dermatology 162, 235 (2010). [18] P. Thomas, L. R. Braathen, M. D€orig, J. Aub€ock, F. Nestle, T. Werfel and H. G. Willert, “Increased Metal Allergy in Patients with Failed Metal-on-Metal Hip Arthroplasty and Peri-Implant T-Lymphocytic Inflammation”, Allergy 64, 1157 (2009). [19] S. Freeman, “Allergic Contact Dermatitis to Titanium in a Pacemaker”, Contact Dermatitis 55, 41 (2006). [20] K. A. Dietrich, F. Mazoochian, B. Summer, M. Reinert, T. Ruzicka and P. Thomas, “Intolerance Reactions to Knee Arthroplasty in Patients with Nickel/Cobalt Allergy and Disappearance of Symptoms after Revision Surgery with Titanium Based Endoprostheses”, J. Deutsche Dermatol. Gesellschaft 7, 410 (2009). [21] Webster’s New Collegiate Dictionary, 11th Edition, Merriam-Webster, Springfield, MA (2003). [22] J. W. Tukey, Exploratory Data Analysis, Addison-Wesley, Reading, MA (1977).