Portable X-ray Fluorescence (PXRF) for CompositionalAnalysis of Early American MetalwareK. Russell and M. Roppolo, Olympus...
of 1

Portable X-ray Fluorescence (PXRF) for Compositional Analysis of Early American Metalware

Handheld XRF analyzers are used worldwide to provide highly specific material chemistry for rapid and accurate identification of alloys and metals. Industrial requirements for quality control, plant maintenance, and profit/loss have prompted portable XRF manufacturers to minimize measurement times, ruggedize analyzer housings, simplify operation, extend the elemental range and detection limits, and improve the accuracy (correctness) and precision (reproducibility) of the results. The nuances of archaeological, conservational, and collectible metalware analyses are similar to those for industry. The objectives of the analysis, the representativeness of the sample measurements, and the condition of the sample are all important considerations. Part of a private early American metalware collection in Salem, Massachusetts, was made available to illustrate the versatility of portable, handheld XRF measurements. Contact us: http://bit.ly/1rDmq94 Sign up for our newsletter: http://bit.ly/1j5FOTy
Published on: Mar 4, 2016
Source: www.slideshare.net


Transcripts - Portable X-ray Fluorescence (PXRF) for Compositional Analysis of Early American Metalware

  • 1. Portable X-ray Fluorescence (PXRF) for CompositionalAnalysis of Early American MetalwareK. Russell and M. Roppolo, Olympus, Waltham, MA USAQualitative and quantitative results, as well as analysis spot images, can be displayed on the DELTA handheld screen or on a the DELTA PC software screen.SAA_2013_03_Poster IntroductionHandheld XRF analyzers are used worldwide to provide highly specific material chemistry for rapid and accurate identification ofalloys and metals. Industrial requirements for quality control, plant maintenance, and profit/loss have prompted portable XRF manufacturersto minimize measurement times, ruggedize analyzer housings, simplify operation, extend the elemental range and detection limits, andimprove the accuracy (correctness) and precision (reproducibility) of the results. The nuances of archaeological, conservational, andcollectible metalware analyses are similar to those for industry. The objectives of the analysis, the representativeness of the samplemeasurements, and the condition of the sample are all important considerations. Part of a private early American metalware collection inSalem, Massachusetts, was made available to illustrate the versatility of portable, handheld XRF measurements.Representative MeasurementsRepresentative measurements are critical because PXRF is a near surface technique with a spot size of approximately 10mm and adepth penetration of about 10 microns for typical alloys. If a piece is very large or thick, several measurements should be made at differentlocations for more representative results. Measurements of small pieces or small sections of a piece may require collimation to reduce theanalyzer spot size further. The condition of a sample is significant because XRF measurements are more accurate and precise for smooth,flat surfaces. If a material is painted, plated, corroded, or has peened surfaces, sample preparation, including grinding, may be requiredfor the most accurate and precise results. The more homogenous a sample is, the more accurate the XRF results will be; fortunately, mostalloys and metals are homogenous.Analytical ObjectivesThe analyst must determine what type of results will meet the objectives – qualitative, semi-quantitative, or quantitative. QualitativeEDXRF results provide the identification of the elements in a given material by observing the keV at which the element’s peaks occur in thespectra. Semi-quantitative results provide relative or comparative amounts of elements in a given material based on peak characteristics(height and/or area). Quantitative results provide the most complete report of sample composition based on peak characteristics and anappropriate calibration curve.ReferencesBrumfield, Gary. 2012. “Iron and Steel in Colonial America.” www.Flintriflesmith.com.Colonial Williamsburg Foundation.LeFever, Gregory. 2007. “Early Pewter Tableware.” Early American Life.Mass, Jennifer, and Catherine Matsen. 2012. “Quantitative non-destructive analysis of historic silveralloys: X-ray fluorescence approaches and challenges.” In Studies in Archaeological Sciences,Handheld XRF for Art and Archaeology, edited by Aaron N. Shugar and Jennifer L. Mass. LeuvenUniversity Press.The Pewter Collectors’ Club of America. 2012. “About Pewter”.www.pewtercollectorsclub.org.Plummer, Don. 1999. Colonial Wrought Iron, the Sorber Collection.Sherman, J. 1955. “The theoretical derivation of fluorescent X-ray intensities from mixtures.”In Spectrochimica Acta (7): 283-306.Smith, Kenneth. 2011. “How Recent XRF Developments Impact Plant Based Alloy Material Testing.”5th Pan American Conference for NDT.Snodgrass, Mary Ellen. 2004. Encyclopedia of Kitchen History. Taylor & Francis Books, Inc.Walton, Steven. 2008. “Iron Beginnings in America”. Medieval Technology and American History.The Pennsylvania State University.IronIron furnaces were constructed in America as early as 1620; and, by 1660, many of the colonies wereproducing iron. Its ore was one of the first loads of material to be shipped back to England. A little morethan 100 years later, the colonies were one of the top global exporters of iron. The Iron Act of 1750, whichrestricted aspects of its manufacture and production, was considered to be a contributor to the start of theRevolutionary War.Three types of iron were used during the colonial period. Cast iron, which is high in carbon and brittle,was used for frying pans, Dutch ovens and canon balls. Wrought iron, which is low in carbon and very tough,was used for anchor chains, nails and musket barrels. Steel, which has varied amounts of carbon dependingon its hardening needs, was used for knife blades, files, saws, springs, musket ramrods and swords. A widelysought type of colonial iron was found in bogs and so named Bog Iron. Dissolved iron in bogs and swampscombined with magnetite and quartz to deposit a material that had a natural resistance to rust due to thecoating caused by its residual silicate content. It was used for tools, iron rails, and cannon balls.The Wrought Iron Trammel is 30” long when extended fully and has 4 slots to vary its length to the firefor different heating temperatures. There are no obvious makers’ marks. Measurements of this piece wererelatively uniform throughout with 99+% iron (Fe), 0.3692-0.4586% cobalt (Co), 0.0557-0.3156% manganese(Mn), and trace amounts of copper (Cu), zinc (Zn), and lead (Pb).Longer testing times and use of a two-beam Rhodium-target tube analyzer would enable the analysisfor silicon (Si), phosphorus (P) and sulfur (S). Handheld XRF cannot measure carbon (C) content. One of thetrammels in the image labeled “Trammels, I-29”, on page 14 of Don Plummer’s Colonial Wrought Iron, theSorber Collection, is very similar to the one analyzed.SilverSilver was expensive and considered a luxury in early America. Much of the silver was imported from England as there were lawsrestricting colonists from importing unfinished silver for manufacture. A colonial silversmith would have been considered an artist whocreated both ornamental and utilitarian pieces, the more common ones being spoons, porringers, tankers, buttons and shoe buckles.Customers took their silver pieces back to the silversmith when they needed repairs or wanted new designs, but did not want to or couldnot purchase more silver. Markings on silver pieces can be attributed to the silversmith and/or the owner. Pure silver, too soft on its own,was alloyed with a base metal to harden it, typically copper. By 1860, a piece marked “sterlingsilver” indicated that it was 92.5% Ag and 7.5% Cu.The Silver Creamer is 10” high and 5” across at the widest part. A decorative pattern isetched around the top, main body, and pedestal. There are no obvious makers’ marks. Thepresence of Pb and gold (Au) in the creamer indicate that it is likely earlier than 1850. Theconcentration of Ag is between 85-88%, indicating that it is likely to be made from Coin Silver,perhaps even prior to the Coinage Act of 1792 which set that Standard at 89.2 wt % Ag.CopperEarly American copper was primarily used to make copper tea kettles for people who could afford tea. It was relatively inexpensive,attractive and conducted heat well. Early kettles were thick and hand hammered; soldering of seams came after 1850. Early copper kettleshad metal handles as they typically hung over a fire to boil; later ones had wood or bone handles to minimize the burning of hands. Copperwas known to react with some contents causing poisonous effects; consequently, many were lined on the inside with tin.The Copper Kettle is 8” high, including the handle, and 4” across at the widest part;the spout is 3” high. There are no obvious makers’ marks. It is thick and appears to haveexperienced significant wear and tear, including blackening. The concentrations of tin (Sn) onthe inside of the lid and on the worn bottom confirm Sn lining. There is higher titanium (Ti), Fe,and Pb content in the spout than the body. The lid contains more nickel (Ni) than the rest ofthe kettle.BrassBrass, an alloy of Cu and Zn, was another type of metalware that early colonists imported from England. Brass was primarily usedfor common things like bells, pins, nails, hinges and candle sticks, but also for navigational and musical instruments and clock parts. Brassmakers could vary the amounts of Cu and Zn to alter the color and softness of the material. Bronze, an alloy of copper and tin, was alsoused, but for pieces that required more strength and weather resistance, like cannons, bells and other metal pieces used out doors. Anearly colonist, who could bring along a large brass kettle, valued it for its beauty, durability and heat conduction.The Large Brass Kettle is 13” high, including the handle, and 9” across at the widestpart; the spout is 6” high. A decorative pattern is etched around the top and main body, butit has no obvious makers’ marks. The brackets vary in composition from the rest of the kettle,with noticeably higher Fe and Pb, and the presence of Sn.Cu Ni Sn Ti Fe PbBody 99.35 0.056 0.044 0.055Handle 99.25 0.067 0.456 0.062 1.166Spout 97.82 0.074 0.157 0.477 0.554 0.531Bottom 97.54 0.048 1.163 0.305 0.474 0.370Lid Inside 83.09 0.452 16.08 0.277 0.037 0.086Lid Outside 99.11 0.584 0.148 0.030 0.092Lid Handle 99.37 0.057 0.083 0.053 0.439Cu Zn Ni Sn Fe PbBody 66.16 33.63 0.057 0.069 0.083Bottom 65.08 34.62 0.058 0.119 0.127Top 65.76 33.99 0.058 0.081 0.111Handle 65.96 33.83 0.059 0.066 0.088Spout 67.08 32.74 0.049 0.059 0.073Lid 67.06 32.80 0.046 0.040 0.054Bracket 1 68.08 27.04 0.053 1.306 0.313 3.206Bracket 2 60.33 35.28 0.137 0.605 0.274 3.375Cu Sn Fe Zr Sb Pb BiHandles 0.0431 74.16 0.1281 0.0284 0.4887 24.95 0.2852Outside Surface 0.0626 35.34 0.0272 5.222 59.07 0.2887Feet 0.1368 40.2 0.1252 0.0262 4.762 53.61 0.4138Ag Cu Pb Zn Au TiBody 85.50 13.67 0.494 0.212 0.091Base 87.54 11.70 0.467 0.210 0.088Handle 87.19 11.18 0.391 0.122 0.106 1.010Lip 87.93 11.51 0.390 0.087 0.081PewterEarly American pewter was an alloy of 84-99% Sn and 1-16% Pb with some Cu, antimony (Sb), and/or bismuth (Bi) that was handspun or mold cast. It was a soft metal with primarily simple designs used by people who could not afford metals like silver (Ag). Towards theend of the 18th century, the use of Pb was minimized for safety, but the resulting alloy did not have the same luster or strength. Althoughmost styles were English, some Americans crafted pewter, especially in Massachusetts, New York, and Connecticut; many used the eagleas their mark.The Pewter Bowl is 3” high and 3” across at the widest part. It is likely a sugar bowl withits wavy edge, two handles and three legs. Marks on the bottom appear to be a crown and rose,typical of an English piece from the 16th century on. Tin was much more expensive than lead andnot as readily available in the colonies; consequently, the very high Pb content (almost 60% in theouter surface) indicates the inferior quality of English export or perhaps a colonist’s rework of existingpewter, stamped with a British mark to increase the perceived value of the piece. This piece is likelyfrom the late 1700s because at that time, wavy edge pieces were popular and inferior quality pieceswere particularly predominant due to British trade laws and sentiments towards the colonies.ConclusionsHandheld XRF analyzers configured for industrial metal and alloys analysis are also ideal tools to perform analyses on a variety ofcollectible metalware. A quick elemental analysis with a standard industrial configuration provides enough information to help determine ifthe composition of a piece falls within the expectations of known authentic pieces. Handheld XRF measurements of light elements or verylow concentrations, such as PPM levels, may require longer test times and/or optional power settings, target material, selectable primarybeam filters, and/or external vacuum/helium or enclosure accessories.Data and spectra can be downloaded directly from the DELTA handheld XRF into Excel for reporting, analysis and averaging. Thesimple, practical averaging method that the unit provides is ideal for samples that are not homogenous.A thorough analysis, with an analyzer configured for all of the elements of interest, such as Si, P, and S, and additional standardsspecifically tailored for the metalware family of interest, could provide even more detailed information when authentication for monetaryvalue and/or conservation is desired. The primary trade-off concerns sample preparation, the analytical value of which must be taken intoconsideration with potentially damaging alterations to the piece.Quantitative AnalysisQuantitative analysis is dependant on a calibration curve which can be stored in the analyzer’s data processing software. Thecalibration curve should be based on an appropriate XRF analysis algorithm because a pure linear relationship between an element’sconcentration and peaks’ characteristics may not exist. Other elements in the material may increase or decrease the excitation of theelement of interest or have peaks that overlap with the peaks of interest. Large differences in the concentrations of the elements that affectone another can further complicate the situation. Overlapping issues are usually handled with peak deconvolution methods stored in theanalyzer’s data processing software.Quantitative alloy analysis is typically based on a Fundamental Parameters (FP) approach which utilizes equations developed byJacob Sherman in the 1950s correlating elemental intensities with sample composition. FP utilizes physical constants and parametersthat are based on the physics of X-rays, incorporating both theoretical and measured intensities for the analyzer’s configuration (tube,detector, filters). This approach relies on one or more certified standard, and is more robust when a standard’s matrix is similar to that ofthe sample and covers the elemental and the concentration ranges of interest. FP is sometimes referred to as “standardless” because themanufacturer has already included the standards’ measurements into the analyzer’s algorithm for the analyst. For most metals and alloysanalyses, FP is the algorithm of choice.Analysis MethodAn Olympus DELTA Handheld X-ray Fluorescence Analyzer, equipped with a single beam tantalum-anode X-ray tube, 40kV/4Wpower, aluminum filter, and large area Silicon Drift Detector was used. A Fundamental Parameters algorithm was employed utilizing puremetal and certified industrial alloy standards. No sample preparation was performed on the collection as per request of the owner. Threemeasurements were taken for 30 seconds each at a given location; and, 1-4 locations were measured on a given area depending on itssize and curvature. Results shown herein are averages.Early American MetalwareThe type of metalware an early American would have used inthe 17th and 18th centuries would have depended on their socio-economic standing. Common items would have included pots andpans, spoons and knives, plates or platters, kettles, candle holders,fireplace implements, weaponry, and tools. The metalware wouldhave included iron, pewter, brass, copper, and silver. There were twobasic ways to create metalware, hammering by hand or casting bymold. Harder metals would have been heated, poured and cast; whilesofter metals would have been shaped and hammered.Recycling metal was common in early America due to financialsituations, conservation practices and English laws that preventedcolonists from manufacturing and selling new pieces. Somehouseholds would have their own marking or initials placed on theirmetalware, especially copper and silver, to discourage others fromtaking it. A large collection would likely also be numbered.www.olympus-ims.com

Related Documents