National Academies Press: OpenBook

Variability of Ignition Furnace Correction Factors (2017)

Chapter: Chapter 3 - Survey of State DOT and Industry Practice Regarding the Use of Ignition Furnaces

« Previous: Chapter 2 - Literature Review
Page 18
Suggested Citation:"Chapter 3 - Survey of State DOT and Industry Practice Regarding the Use of Ignition Furnaces." National Academies of Sciences, Engineering, and Medicine. 2017. Variability of Ignition Furnace Correction Factors. Washington, DC: The National Academies Press. doi: 10.17226/24707.
×
Page 18
Page 19
Suggested Citation:"Chapter 3 - Survey of State DOT and Industry Practice Regarding the Use of Ignition Furnaces." National Academies of Sciences, Engineering, and Medicine. 2017. Variability of Ignition Furnace Correction Factors. Washington, DC: The National Academies Press. doi: 10.17226/24707.
×
Page 19
Page 20
Suggested Citation:"Chapter 3 - Survey of State DOT and Industry Practice Regarding the Use of Ignition Furnaces." National Academies of Sciences, Engineering, and Medicine. 2017. Variability of Ignition Furnace Correction Factors. Washington, DC: The National Academies Press. doi: 10.17226/24707.
×
Page 20
Page 21
Suggested Citation:"Chapter 3 - Survey of State DOT and Industry Practice Regarding the Use of Ignition Furnaces." National Academies of Sciences, Engineering, and Medicine. 2017. Variability of Ignition Furnace Correction Factors. Washington, DC: The National Academies Press. doi: 10.17226/24707.
×
Page 21
Page 22
Suggested Citation:"Chapter 3 - Survey of State DOT and Industry Practice Regarding the Use of Ignition Furnaces." National Academies of Sciences, Engineering, and Medicine. 2017. Variability of Ignition Furnace Correction Factors. Washington, DC: The National Academies Press. doi: 10.17226/24707.
×
Page 22
Page 23
Suggested Citation:"Chapter 3 - Survey of State DOT and Industry Practice Regarding the Use of Ignition Furnaces." National Academies of Sciences, Engineering, and Medicine. 2017. Variability of Ignition Furnace Correction Factors. Washington, DC: The National Academies Press. doi: 10.17226/24707.
×
Page 23
Page 24
Suggested Citation:"Chapter 3 - Survey of State DOT and Industry Practice Regarding the Use of Ignition Furnaces." National Academies of Sciences, Engineering, and Medicine. 2017. Variability of Ignition Furnace Correction Factors. Washington, DC: The National Academies Press. doi: 10.17226/24707.
×
Page 24
Page 25
Suggested Citation:"Chapter 3 - Survey of State DOT and Industry Practice Regarding the Use of Ignition Furnaces." National Academies of Sciences, Engineering, and Medicine. 2017. Variability of Ignition Furnace Correction Factors. Washington, DC: The National Academies Press. doi: 10.17226/24707.
×
Page 25
Page 26
Suggested Citation:"Chapter 3 - Survey of State DOT and Industry Practice Regarding the Use of Ignition Furnaces." National Academies of Sciences, Engineering, and Medicine. 2017. Variability of Ignition Furnace Correction Factors. Washington, DC: The National Academies Press. doi: 10.17226/24707.
×
Page 26

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

18 3.1 Introduction A survey was developed to help focus the NCHRP Project 9-56 research (see Appendix A). A list of contacts for state and provincial highway agencies and the FHWA Federal Lands Divisions was compiled, and potential contacts were con- firmed by e-mail prior to sending the survey. A total of 105 nearly complete responses were received. Additional largely incomplete responses were deleted. There were a total of 60 agency responses representing 42 of 50 U.S. states, seven of 10 Canadian provinces, and the Federal Lands Divisions. Some agencies submitted more than one response. There were an additional 37 responses from contractors, seven responses from testing laboratories, and one unclassifiable response. 3.1.1 Furnace Type and Age The survey respondents reported operating a total of over 365 furnaces. Figure 13 indicates the types of furnaces in use. Respondents could indicate more than one brand/model, so the percentages exceed 100%. The vast majority of the fur- naces used (93.3%) included internal balances. The major- ity of survey respondents (56.3%) indicated differences in correction factors for asphalt content determinations with different brands, models, or locations of ignition furnaces performing the test. Furnace age likely affects maintenance requirements. Fig- ure 14 shows the ages of the furnaces operated by the respon- dents. Respondents could pick more than one age range, so the percentages again exceed 100%. The range of furnace ages appears to be normally distributed, with a median age of 5 to 10 years. 3.1.2 Test Procedure Specified Survey respondents were asked to identify the test proce- dure they used for asphalt content determination. AASHTO T 308 was the most common test method and was indicated by 78.5% of those answering the question. ASTM D6307 was used by 21.5% of the respondents. Several agencies reported their own test methods. In many cases, these represent slight variations from either AASHTO T 308 or ASTM D6307. The differences are summarized in Appendix B. 3.1.3 Ignition Furnace Calibration A primary objective of this study was to improve methods for calibrating ignition furnaces for asphalt content determi- nation. As noted previously, 56.3% of respondents indicated differences in correction factors for different brands, models, or locations of furnaces. Figure 15 indicates the factors identified by respondents as affecting the ignition furnace calibration for asphalt content determination. Early in the development of the test procedure, researchers realized that each aggregate source tended to require a unique correction factor. This cor- rection factor may account for aggregate loss or loss of inter- nal aggregate moisture not removed by normal drying. This significant effect of aggregate type/geology was identified by 92.2% of the 90 respondents who answered the question. Test temperature was identified by 37.8% of respondents as affecting aggregate correction factors. Several agencies and researchers have investigated lower test temperatures to reduce aggregate breakdown/loss during the test. Asphalt content was identified by 21.1% of respondents as affecting the cali- bration factor. Early calibration procedures used samples at optimum and ±0.5% of optimum for calibration. This was later simplified to only include samples at optimum asphalt content. Test temperature tends to increase when the asphalt sample ignites. Samples with higher asphalt content or larger samples have more liquid asphalt to burn and may result in higher peak test temperatures. Only 14.4% of survey respondents indicated that hydrated lime affects ignition furnace correction factors. This com- pares to 34.8% of respondents (92 answered the question) who say they use hydrated lime in asphalt mixtures. C h a p t e r 3 Survey of State DOT and Industry Practice Regarding the Use of Ignition Furnaces

19 68.5% 20.2% 5.6% 18.0% 6.7% 2.2% 0% 10% 20% 30% 40% 50% 60% 70% 80% Thermolyne Series 859/945 Fisher Thermo/Thermolyne Series 1087/1275 Troxler 4155 Asphalt Analyzer Troxler 4730/4731 NTO Gilson HM-378 Binder Ignition System Carbolite Asphalt Binder Analyzer Figure 13. Brands/models of ignition furnaces used by survey respondents. 92.2% 37.8% 21.1% 14.4% 0% 20% 40% 60% 80% 100% Aggregate type/geology Test temperature Asphalt content Inclusion of hydrated lime Figure 15. Factors affecting ignition furnace calibration for measured asphalt content. 26.4% 37.4% 51.7% 38.5% 18.7% 0% 10% 20% 30% 40% 50% 60% < 2 years old 2 - 5 years old 5 - 10 years old 10 - 15 years old > 15 years old Figure 14. Ages of furnaces operated by survey respondents.

20 Additional factors were identified in the question’s com- ment section. Two respondents indicated RAP or RAS as factors affecting the correction factor. One agency added (1) length of vent pipe, (2) cleanliness of furnace chamber, and (3) how the baskets are loaded with mixture as being factors. Another agency mentioned the technician’s atten- tion to detail when preparing the calibration samples. Figure 16 shows the typical range of ignition furnace asphalt content correction factors. The majority (89.9%) of respondents indicated typical correction factors of less than 1.0%. Figure 17 shows a wide range of aggregate types used by the respondents. Granite, gravel, and limestone are the most prominent aggregate types. As noted previously, 56.3% of 80 respondents indicated that they see differences in correction factors determined for different brands, models, or locations of furnaces. Some agencies allow correction factors to be shared (e.g., the same correction factor is used by the contractor and the agency). A unique correction factor for each furnace was required by 72.0% of 82 respondents. Some specific comments from agencies that share calibration factors include: • Connecticut DOT typically uses the correction factor developed by the contractor. If in question, it will perform using its own. • West Virginia DOT uses the contractor’s correction fac- tor for normal QC/QA, but each laboratory is required to develop its own when using new percent-within-limits specifications. • Oklahoma DOT uses the correction factor determined by the contractor during mix design. If a furnace other than a Thermolyne or Thermo/Fischer Scientific is used, calibra- tion should be verified or determined for that furnace. One contractor noted that it had shared among its own furnaces in the past but noticed some differences with older models. It was working toward the use of furnace-specific factors. 67.4% 49.4% 6.7% 3.4% 0% 10% 20% 30% 40% 50% 60% 70% 80% <= 0.50 0.51 - 1.00 1.00 - 2.00 > 2.00 Figure 16. Typical asphalt content correction factor range. 26.4% 11.0% 34.1% 64.8% 60.4% 57.1% 29.7% 38.5% 18.7% 0% 10% 20% 30% 40% 50% 60% 70% Basalt Diabase Dolomite Granite Gravel Limestone Quartzite Sandstone Slag Figure 17. Aggregate types used in respondents’ asphalt mixtures.

21 A concern identified prior to the survey was the valid- ity or quality of the calibration samples. The furnace cali- bration can only be as accurate as the calibration sample. The research team has previously observed errors in the calibration factor resulting from improperly “buttering” the mixing bowl used to prepare the sample. Essentially, some asphalt and fines will be left in the mixing bowl and on the mixing utensils. This can vary from mixture to mixture; for example, stone matrix asphalt with a polymer-modified binder may leave more material than a dense-graded mixture with neat asphalt. The party preparing the calibration samples has responsi- bility for their accuracy. Figure 18 summarizes responses from 86 respondents who answered the question, “Who prepares correction factor samples?” The most common preparer is the contractor (46.5%). The agency and contractor each prepared their own samples in 37.2% of cases. Survey respondents were asked how they ensured that ignition furnace correction factors were representative of the design. Answers were by comment only, and 72 responses were received. This question was added to the survey in response to a concern by one of the advisory panel members. While not directly stated, it seems the concern was whether the correct amount of asphalt was added to the correction fac- tor and particularly that a lesser amount, which would tend to result in a smaller correction factor, was not used. The majority of respondents stated that they carefully batched to the design gradation. Two respondents indicated that they try to match plant conditions as closely as possible. Some use cold-feed belt cuts to obtain aggregate samples. The following responses offer some alternative ideas. • Arizona DOT indicated that it adjusted correction factors to plant asphalt tank sticks made over the first 5 days of pro- duction if the difference between the ignition furnace tests and asphalt content from plant sticks was more than 0.1%. • Missouri DOT monitors mix volumetric properties and theoretical maximum gravity to make sure they align with the design at the reported asphalt content. • Three respondents indicated that they check with solvent extractions. • A number of agencies conduct spot checks or otherwise check based on experience and judgment. There is some evidence that ignition furnace correction factors can change with time as different ledges or areas of a quarry are processed. Figure 19 indicates that many respon- dents (48.8%) determine correction factors once a year. How- ever, 18.6% of respondents indicated that correction factors were reevaluated more than once per year, and 17.4% indi- cated that they were never reevaluated. While the majority of respondents indicated that cor- rection factors were reevaluated at least once a year, a follow-up question asked if they had criteria to trigger a new correction factor. Only 38.4% of respondents indicated that they had criteria to trigger a reevaluation. A number of the respondents indicated that they use engineering judgment to decide when it is necessary to reevaluate correction factors, considering agency results differing from producer results, erratic results, and comparison with volumetric properties or maximum specific gravities. Some specific triggers for reevaluating calibration factors that were listed included those discussed in the following. • Reevaluating when new material has been crushed for the same mixture. • Three agencies mentioned verification by solvent extrac- tion. An additional agency uses nuclear gauge comparisons and the plant’s daily asphalt usage report. One agency has tolerances to compare to asphalt tank sticks. Figure 18. Entity preparing ignition furnace correction factors.

22 • Minnesota DOT reevaluates calibration factors if there are consecutive tolerance issues between agency and producer, when aggregate percentages change by more than 10%, or when there are changes in limestone or dolostone content. • Connecticut DOT requires a new correction factor when the RAP proportion changes by more than 5% from the job mix formula, when the binder content in the processed RAP changes by more than 2%, or when the theoretical maximum specific gravity changes by 0.06. • South Carolina DOT requires monthly verifications. • Montana DOT recalibrates each time it moves its mobile trailers. As shown in Figure 16, the majority of correction factors for asphalt content are less than 1%. Some agencies modify test procedures when correction factors become too high. AASHTO T 308 recommends lowering the test temperature from 1,000°F (538°C) to 900°F (482°C) when correction factors exceed 1.0%. Based on 85 responses, only 20.0% of respondents modify test procedures when correction factors are inconsistent or high. Ten agencies reported lowering the test temperature when correction factors exceed 1.0%. Two contractors reported lowering the test temperature when cor- rection factors exceed 0.5%. Minnesota lowers test tempera- ture when the mixture’s limestone/dolostone content exceeds 20%. One agency reported switching to Troxler NTO, and another changes burn profiles with Troxler NTO when cor- rection factors exceed 1.0%. Arizona DOT’s test procedure allows samples to be tested at as low as 800°F (427°C). AASHTO T 308, Method A, specifies that the test is com- plete when the mass loss does not exceed 0.01% for 3 consecu- tive minutes. This end point definition can also be changed for mixtures with high correction factors. AASHTO T 308 notes using 0.02%. Oregon DOT uses 0.03% for testing reclaimed shingles. 3.1.4 RAP and RAS Use of RAP and RAS is increasing. Twenty-nine of 57 respon- dents (50.9%) indicated that they did not have specific issues when calibrating the ignition furnace with mixtures contain- ing RAP or RAS. Eight respondents indicated concerns related to variability of asphalt content and gradation with mixtures containing RAP or RAS. Eight respondents indicated diffi- culty determining accurate asphalt contents with RAP or RAS. Specific comments from respondents included those in the following. • Calibrate with virgin aggregate only to obtain a more accu- rate correction factor. • Assume a correction factor of 0.5% for RAP. • Assume a correction factor based on local aggregates for RAP. RAS asphalt content by ignition test is generally 25% higher than by solvent extraction. • The correction factor used is the same as for a virgin mixture (do not use RAS). • Not getting enough asphalt in mixtures. Correction factors not accounting for RAP properly. • One respondent does both solvent extractions and ignition furnace tests. Extractions produce lower asphalt contents. • If RAP contains asphalt rubber, undigested rubber will burn off during ignition but is not considered as binder in a solvent extraction. • Two respondents indicated increasing correction factors with increased RAP contents. Two respondents indicated that RAP or RAS does not always burn clean. Figure 19. Frequency at which correction factors are determined or reevaluated.

23 • One respondent noted that the sample size for RAS must be small, approximately 200 g. • One respondent indicated problems with consultant labo- ratories that were unfamiliar with procedures or not careful in weighing correction samples. 3.1.5 Handling Procedures During Testing Once a correction factor is determined, certain practices during routine testing may affect variability. The questions discussed in the following were designed to gather more infor- mation on these practices. The Thermolyne and Thermo/ Fischer furnaces use a temperature compensation factor in addition to the correction factor for aggregate loss. The tem- perature compensation factor addresses apparent mass changes that occur as room-temperature baskets heat up to test tem- perature. Asphalt mixtures are hot when sampled and split, and depending on work flow, samples are often loaded hot, immediately after splitting. However, depending on the vol- ume of testing and number of furnaces a laboratory has, samples may cool in the baskets prior to testing. Figure 20 indicates that most users load hot samples, and 20% of users load either hot or room-temperature samples, depending on circumstances. Respondents were asked to identify typical burn times. Sam- ple size, furnace temperature, and total asphalt content of the mixture all affect burn time. Figure 21 shows the distribution of observed test times. The vast majority of samples burn for between 30 min and 1 h. Both AASHTO T 308 and ASTM D6307 include methods for measuring asphalt content using either the furnace’s inter- nal balance (Method A) or by weighing on an external balance (Method B). As discussed in Section 3.1.1, the vast majority of survey respondents use furnaces with internal balances. How- ever, not all respondents use the internal balance to determine the sample mass loss and, hence, calculate the sample asphalt 20.0% 12.9% 67.1% 0% 10% 20% 30% 40% 50% 60% 70% 80% Either Room temperature Hot Figure 20. Sample temperature when loaded into furnace. Figure 21. Typical sample burn times.

24 content. While 93.3% of respondents reported using furnaces with internal balances, only 62.8% of 86 respondents answer- ing the question indicated that they use the internal balance to calculate asphalt content. The remainder (37.2%) weigh the sample on an external balance. 3.1.6 Gradation Analysis The ignition furnace test method produces a post-plant- mixing aggregate sample for gradation analysis. Gradation analysis is performed on the recovered aggregate by 95.5% of the survey respondents. For some aggregate types, the ignition test can apparently alter the recovered gradation. Some users suggest that fines do not stick to coarse aggregate particles post-ignition and therefore do not require washed gradations to be run. Washed gradations are run on aggregate recovered using the ignition furnace by 86.2% of the survey respondents. Early in the use of the ignition furnace, a higher percent passing the No. 200 sieve was observed compared to solvent extractions. This was later traced to a difference in washing technique. The laboratory using the ignition furnace used an aggregate drum washer; the laboratory performing solvent extractions washed the recovered aggregate samples by hand. Figure 22 shows the distribution of hand washing, using an aggregate drum washer, and doing both. The AASHTO Materi- als Reference Laboratory now requires accredited lab oratories to calibrate the automatic drum washer time to produce an equivalent percent passing the No. 200 sieve as hand washing. Correction factors can also be developed for the recovered aggregate gradation and are used by 33.3% of respondents. A number of agencies indicated that they did not see a need for this correction factor or that the need to correct grada- tions was rare. • Seven respondents indicated that they use the procedure in the annex of AASHTO T 308. • Ontario uses a similar procedure but develops a correction factor if the difference exceeds 1.0% for sieves larger than No. 200 and 0.3% for No. 200 sieves. • Three agencies develop correction factors only for the per- cent passing the No. 200 sieve. • West Virginia only uses a gradation correction factor for percent-within-limits contracts. 3.1.7 Furnace Installation and Maintenance Ignition furnaces use a fan to pull air into the furnace cham- ber to support ignition and provide negative pressure to min- imize escape of emissions into the laboratory. The manner in which a furnace is connected to exhaust ductwork and the maintenance of the fan likely affect furnace operation. Only 37.2% of respondents indicated that they had procedures for installing and maintaining their furnaces. Twenty-seven respondents provided comments. Nine of these indicated that they follow the manufacturer’s instructions. The Thermolyne and Thermo/Fischer furnaces encourage a lift test. Air is drawn in under the balance plate. If the scale is zeroed prior to starting the fan, turning the fan on creates lift, resulting in a negative scale reading. The manufacturer pro- vides guidelines for what this lift should be to provide adequate air flow through the furnace. Six respondents indicated that they performed the lift test with a frequency ranging from daily to every 3 months. Six respondents indicated that they clean the fan/ductwork, most on a monthly basis. A number indicated that they oil the fan. Arizona and Connecticut have written maintenance procedures. Arizona’s does not address oiling the fan but includes the other previously described items. Connecticut adds adjusting the door latch. British Columbia indicated that it adds checking to see that all of the heating elements are working properly and checking the door for leaks or other smoking problems that might indi- cate exhaust entering the laboratory. Figure 22. Method for performing washed gradation.

25 The length of the exhaust duct likely affects air flow through the furnace and may affect the sample burn. Reported duct lengths ranged from 1.5 to 40 ft. The average duct length was just over 10 ft. Some agencies tie multiple furnaces to a single, larger duct system. Some users vent furnaces into a fume hood or larger exhaust hood with its own fan. More than one furnace is tied to the same ductwork by 13.1% of survey respondents. Ninety-degree elbows significantly reduce air flow. Respondents were asked how many of these they had in their ductwork, and several noted that they avoided 90-degree elbows. The average number of elbows was slightly less than one, and the maximum reported was three. Several respon- dents indicated that the number of elbows varied depending on the installation. Two respondents indicated that they used two 45-degree elbows for improved air flow. Many early fur- naces came with a flexible, stainless-steel exhaust pipe that did not allow sharp bends. A final question designed to address the need for mainte- nance and cleaning was the average number of samples tested in each furnace per year. The responses ranged from two to 1,200 samples per year, with an average of 285. Figure 23 shows a distribution of the results. 3.2 Summary of Survey of State DOT and Industry Practice A survey was conducted to help the researchers refine the experimental plan aimed at identifying factors that influ- ence ignition furnace correction factors and variability and developing guidelines for installation and maintenance. The survey addressed furnace type and age, test procedures used, correction factors, RAP and RAS, handling procedures during testing, gradation analysis, and installation and maintenance. The vast majority of respondents use furnaces with inter- nal balances. However, 37.2% of respondents calculate the asphalt content based on masses measured on an external scale. Furnace age is approximately normally distributed from new to over 15 years old, with a median age of 5 to 10 years. The length of exhaust ductwork connected to a furnace varied from 1.5 to 40 ft, with an average of approximately 10 ft. Air flow is restricted by 90-degree elbows. While many respon- dents avoid these where possible, up to three were used, with an average of slightly less than one. Some furnaces are vented into fume/exhaust hoods. Thirteen percent of respondents hooked multiple furnaces to the same ductwork. Most respondents use AASHTO T 308 or an agency mod- ification thereof. Almost all respondents identified aggre- gate geology as affecting the correction factor for measured asphalt content. Differences in correction factors were seen by 56.3% of respondents for different brands, models, or locations of furnaces. A unique correction factor for each furnace was required by 72% of respondents. This unique correction factor could account for differences in installa- tion, such as duct length or number of elbows or differences in use and maintenance, even for the same brand and model of furnace. Several agencies, however, reported good success with sharing correction factors. Correction factors are most commonly determined on a yearly basis. Most correction factors for measured asphalt con- tent are less than 1.0%. AASHTO T 308 recommends lowering the test temperature to 900°F (482°C) for correction factors exceeding 1.0%. Some agencies allow temperatures as low as 800°F (427°C). Other agencies alter the definition of the end point (e.g., the allowable percentage of mass change over a period of time) to reduce the variability and magnitude of cor- rection factors. Sample preparation may be a major factor in correction factor accuracy and variability. The use of mixtures with RAP and RAS does not seem to pose significant problems with the ignition furnace, although a number of agencies indi- cated concerns over potential variability of RAP stockpiles. The vast majority of respondents run gradation analyses, primarily washed, on the recovered aggregate. Samples are Figure 23. Number of samples tested per year.

26 both hand washed and washed with aggregate drum washers. Only a third of respondents use correction factors for the recovered gradation. Most gradation correction factors are based on AASHTO T 308. Near the conclusion of the survey, respondents were asked to identify areas where they had concerns with the test pro- Figure 24. Significant problems observed with test. cedure. Fifty-six of 106 respondents answered the question. The responses are summarized in Figure 24. Respondents could choose more than one response, so items total more than 100%. The majority of respondents volunteered to participate in round-robin testing to help improve the test procedure.

Next: Chapter 4 - Experimental Plan Description »
Variability of Ignition Furnace Correction Factors Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB's National Cooperative Highway Research Program (NCHRP) Research Report 847: Variability of Ignition Furnace Correction Factors explores the significant influences that affect the variability of asphalt and aggregate correction factors for ignition furnaces. The report presents a proposed practice in American Association of State Highway and Transportation Officials (AASHTO) standard format for installation, operation, and maintenance of ignition furnaces to minimize the variability in correction factors between furnaces.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!