1.1 Evaluation of the Analytical Method for Measuring Occupational

Exposures during Soldering and Brazing Operations

The main concern of this evaluation is to determine whether this instrument is able to detect, correctly identify, accurately and precisely measure exposures near the respective PELs. The general procedure for the analysis of samples by an ARL ICP has been previously described (11.2) and evaluated (11.3). This validation examines the analytical response of an ARL spectrometer to eight elements commonly found in soldering and brazing operations. The eight elements (Ag, Be, Cd, Cu, Pb, Sb, Sn, and Zn) have been validated using concentration levels of approximately 0.5, 1, and 2 times the Transitional or Final Rule Permissible Exposure Limits (PELs) (11.4) as Time Weighted Averages (TWAs).

[Note: After the evaluation, the regulation for exposure to cadmium was changed to 0.005 mg/m³ as a TWA. Please see the Code of Federal Regulations (29 CFR 1910.1027 or 29 CFR 1926.63) for further information.]

In addition, this method is sensitive enough to determine the Short-Term Exposure Limit (STEL) and Ceiling concentrations for Be. This procedure does not differentiate between dusts and fumes.

Spiked samples were used to evaluate the method. Each spike was calculated to represent the amount of analyte found at a concentration near its respective PEL when taking a 400- to 500-L sample. This air volume range is approximately equivalent to taking 4-h samples when collected at a flow rate of 2 L/min.

1.2 Spectrometer Description

The ARL spectrometer includes a 1.0-m path length with 1,080 grooves/mm grating and a solid-state radio frequency (RF) generator. Standard operating conditions for this instrument were developed through a series of benchmark tests recommended for most analytical applications. Equipment and operating conditions used during this evaluation are listed in Table 1. This 36-channel instrument was configured for the simultaneous determination of 35 elements, as listed within the line library in Table 2. Two channels are used for Fe; one wavelength (259.94 nm) is appropriate for dilute concentrations, while the other (271.44 nm) is used for quantitating higher concentrations of Fe. The line library contains information for the line array wavelengths and spectral orders as set up by the instrument manufacturer. The resolution at a specific wavelength is dependent upon the grating and spectral-order of the line.

Spectral resolution is defined as full width at half maximum (FWHM), i.e., at half the height, the measured width of the peak signal. Using the grating specified in Table 1, the resolution/spectral order relationship is described below:

Table of Order vs. Spectral Resolution

Order	Wavelength Range (nm)*	Resolution (nm)	SAMI Range (motor steps)
1st	400-800	0.046	80
2nd	235-400	0.023
3rd	165-235	0.015**
* wavelength range is general - slight overlaps into the next order are possible ** third order provides the best resolution

The SAMI is a spectral scanning device that is further described in Section 3.1.

As shown above, a second-order peak has a resolution of 0.023 nm. A ±80 motor step scan of a second-order peak would easily cover a range of 0.046 nm.

The analysis of additional elements beyond the eight validated allows for a more complete spectral interference correction and provides for additional screening capability. A list of the elements screened and validated is provided in Table 2 and further displayed as calibration standards in Table 3.

1.3 Standard and Reagent Purity

a) Standards:

Standards used during this study were prepared in a 32% HCl/4% HNO₃ matrix except as noted in the text. Single and multi-element standards were used to evaluate the instrument. Chemical compatibility of the constituents was taken into account in order to avoid precipitation when preparing the multi-element standard solutions.

Two multi-element standards and a reagent blank, as listed in Table 3, were used for instrument calibration of the validated elements. The reagent blank consisted of a 32% HCl/4% HNO₃ mixture. A third standard solution (STD SOLN 3) is used to calibrate the screened elements.
 

---

Note: The concentration and combination of elements in each standard mixture was selected to minimize interferences for the particular instrument and wavelengths used. Analysts using instruments with different wavelengths/elements than stated may have to use alternate mixtures.

---

All standard solutions used for instrument calibration (Table 3) or spectral interference determinations were prepared by serial dilution.

b) Chemicals Used for Validation:

Sample Digestions and Dilutions

Nitric Acid (HNO₃, 69.0 - 71.0% w/w) and Hydrochloric Acid (HCl, 36.5 - 38.0% w/w) ("Baker Analyzed" Reagents, J.T. Baker Chemical Co., Phillipsburg, NJ).

Instrument Calibration

ICP Standard Solutions (1,000 µg/mL) ("Instra-Analyzed" Atomic Spectral Standards, J.T. Baker Chemical Co.)

Filter Spiking

Most solutions used for filter spiking were prepared from standards obtained from Inorganic Ventures, Inc. (Brick, NJ). In addition, ICP standard solutions (10,000 µg/mL) from Specpure (Aesar/Johnson Matthey Inc., Seabrook, NH) were used for any high concentration spikes.

Inter-element Corrections

Inorganic Venture and J.T. Baker standards were primarily used to determine inter-element corrections.

1.4 Instrument Calibration

The polychromator was calibrated according to the manufacturer's procedures. The reagent blank and the multi-element calibration solutions (Table 3) were matched to the sample matrix of 32% HCl/4% HNO₃.

During the calibration process, linear regression coefficients were calculated by the Digital Equipment Corporation (DEC) 11/53 computer from standard response readings. All calibrations were determined by first-order regression only.

1.5 Soldering and Brazing Discussion

This validation primarily addresses air samples collected during soldering or brazing operations. The eight elements most commonly found in these operations were: Ag, Be, Cd, Cu, Pb, Sb, Sn, and Zn. Other elements can be added to this list. The capability for expanding the analysis to other elements is mainly dependent on laboratory instrumentation, element solubility, and stability in the acid matrix used for digestion.

Applicable OSHA Permissible Exposure Limits (PELs) (11.4) are listed in Table 4. This Table lists both the Transitional and Final Rule PELs for substances which may be present in solder and brazing operations. The Transitional and Final Rule Limits are identical with the exception of a Final Rule Ceiling Limit for Be, a TWA for tin oxide, and additional STELs for zinc chloride and zinc oxide.

1.5.1 Welding, Soldering, and Brazing Applications (11.5, 11.6)

According to Kirk-Othmer (11.5), welding, brazing, and soldering are metal-to-metal bonding processes. The temperature at which the joint is made is the primary feature which differentiates between welding, soldering, or brazing. Solder alloys melt below 800°F (427°C) and brazing alloys melt above 800°F.

a) Welding:

According to Kirk-Othmer (11.5), in welding operations, similar components are fusion-bonded at or just below their melting points. The filler metal is either puddled into relatively wide gaps, or the metal surfaces being joined are partially melted and bonded by fusion or by a combination of puddling and fusion.

b) Brazing and soldering:

In most brazing and in almost all soldering operations, the components are molecularly bonded well below their melting points. An exception is the brazing of aluminum and magnesium alloys. Generally, both soldering and brazing involve the introduction of a non-ferrous filler material. While similar in concept, brazing and soldering are not identical. The temperatures at which the alloys melt provide the primary difference between the two procedures.

The major brazing filler metals are copper, brass, bronze, and silver alloys. The filler metals are drawn into closely fitted joints by capillary action and they bond and solidify without melting the components. Zinc and cadmium can volatilize from zinc- and cadmium-containing brazing alloys during brazing.

1.5.2 Specific Components of Solders

The composition and use of some common solders is shown in Table 5. A more detailed description of some of the common solder alloys is given below:

a) Lead/tin:

Previously, most solder alloys were composed of combinations of tin and lead.

b) "Lead-free":

Recently, "lead-free" solders have become more prevalent. A sample of this type of solder was obtained and a qualitative analysis by ICP and X-ray fluorescence (XRF) procedures at the SLTC confirmed this solder contained an alloy of tin, silver, copper, and bismuth. The "lead-free" solders have been reported as possessing characteristics similar to "50 Tin/50 Lead" solders, and a greater tensile and shear strength than those made with "50 Tin/50 Lead" or "95 Tin/5 Antimony" solders. "Lead-free" solders have been substituted for lead solders in plumbing applications, but have not been noticeably used in the electronics industry. This lack of use in the electronics industry is apparently due to the high temperature needed to achieve the "lead free" solder's melting point.

c) Silver:

Many silver solders contain cadmium in varying amounts.

d) Antimony/tin:

These solders are also used as brazing alloys.

e) Cadmium/silver:

These are used with much higher temperatures and are suitable for use on copper and aluminum. This solder will also produce a very high tensile strength.

f) Zinc:

Many of the alloyed solders may contain varying amounts of zinc.

g) Indium:

This is used for special applications, i.e., adhering glass to glass or glass to metal. The low vapor pressure of these solders makes them useful for seals in vacuum systems.

h) Other metals:

Other trace contaminants present in base and filler metals include arsenic, chromium, bismuth, cobalt, nickel, selenium, thallium, and vanadium.

1.5.3 Solder and Brazing Health Hazards (11.5, 11.6)

Solder's greatest danger to health lies in the presence of lead or cadmium in the solder alloy. Historically, lead was present in large amounts in most solder alloys, and some special alloys contained cadmium.

In the past, the four most hazardous metals commonly found during soldering and brazing processes were lead, cadmium, beryllium, and zinc. More recent solder formulations attempt to exclude most of these elements. Some of the potential symptoms and hazards incurred from exposure to these and other elements are listed below (11.6) and current PELs for validated elements are listed in Table 4.

a) Lead:

Lead is used in the soldering process in the form of lead/tin and lead/silver filler metals. When heated, lead oxide fumes are formed. Excessive exposure to lead oxide fumes can result in lead poisoning. Symptoms include loss of appetite, indigestion, nausea, vomiting, constipation, headache, abdominal cramps, nervousness, and insomnia. According to Kirk-Othmer (11.5), lead is absorbed through the mucous membranes of the lung, stomach, or intestines and then enters the bloodstream.

b) Cadmium:

Cadmium is found in some silver and zinc solders, and in some base metals. When heated, cadmium oxide fumes can be generated. Excessive exposure to these fumes can result in cadmium poisoning, symptoms of which include dry cough, irritation of the throat and nasal passages, ulceration of the nose, tightness of chest, restlessness, and renal damage. Cadmium is a suspected carcinogen and a higher incidence of prostate and lung cancers are noted among workers in occupations that use cadmium in their processes.

c) Beryllium:

Beryllium is used in magnesium filler metals for furnace brazing and in some aluminum brazing filler metals. While soldering, temperatures are normally too low to generate fumes from beryllium; however, the heat involved in brazing can generate beryllium fumes, which are extremely hazardous. Short-term exposure to these fumes may result in a chemical pneumonia. Long-term effects include shortness of breath, chronic cough, loss of weight, and fatigue. Beryllium is a suspect human carcinogen.

d) Zinc:

Zinc is used in large amounts in zinc-cadmium and zinc-aluminum solders and in some base metals. When heated, zinc oxide fumes are generated. Excessive exposure to freshly formed zinc oxide fumes can result in an illness called metal-fume fever or "zinc chills." Symptoms include the presence of a sweetish or metallic taste in the mouth, dryness and irritation of the throat, coughing, a feeling of weakness, fatigue, and a general malaise condition similar to the flu.

According to Kirk-Othmer (11.5), zinc or tin chlorides are found in some fluxes.

e) Antimony/tin:

The potential health hazard is moderate because harmful amounts of antimony or tin fumes are not generally formed.

f) Indium:

Although human exposures concerning contact with indium or its compounds have not been reported, animal studies indicate significant lung impairment from respiratory exposures.

g) Other Metals (11.6):

Other trace metals present in base and filler metals which can give off toxic fumes include arsenic, chromium, bismuth, cobalt, nickel, selenium, thallium, and vanadium. It should be noted that a specific PEL has not been assigned to bismuth at this time. NIOSH has stated that arsenic is a suspected lung and lymphatic carcinogen, and hexavalent chromium is a suspected lung carcinogen. The amount of fumes generated from these trace metals is usually small, and hazardous concentrations are not normally found in these operations. Soldering and brazing with filler or base metals containing these trace elements should be conducted in well-ventilated areas to be certain that hazardous concentrations do not exist. Compounds of these metals may also be present.

1.5.4 Selection of Solder/Brazing Elements for Method Evaluation

The elements to be evaluated for analysis by ICP were selected based upon several factors which included:

1. The severity of potential health hazards from soldering and brazing operations.
   
2. Probability of occurrence of those hazards in the workplace environment.
   
3. The instrumental considerations including compatibility of the selected analytes.
   
4. The applicability of a dissolution procedure for the sample matrix.

For example, due to its limited use in industry, indium is not a validated element in this report. Analysis can be performed for indium by using OSHA method no. ID-121. An alternate ICP procedure (11.2, 11.7) is available which can determine the elements Be, Cd, Cu, Pb, Sb, and Zn. This alternate procedure is unable to accurately quantitate the elements Ag and Sn due to solubility problems from the H₂S0₄/HCl digestion used.

1. Investigation and correction for spectral line interferences.
   
2. Determination of over-all analytical precision and accuracy. Analyses of 18 samples for each element (6 samples at each test level) except for Ag where 3 samples were determined at each test level.
   
3. Determination of analytical detection limits.
   
4. Determination of working ranges for the elements validated.
   
5. Evaluation of spiked Quality Control (QC) samples.
   
6. Discussion of the digestion procedure.
   
7. Determination of a standard reference material.
   
8. Determination using alternate wavelengths.
   
9. Summary.

3.1 Procedure

Inter-element corrections for elements likely to be found in soldering and brazing operations were experimentally determined by identifying and then evaluating the magnitude of each interference on each analytical line. The interferences were first qualitatively identified through peak scans and then a quantitative correction factor was calculated. Inter-element corrections were experimentally determined by identifying and then evaluating the magnitude of the interferences as stated below.

The identification of interferences was first determined by scanning peak profiles. Three types of peak scans were conducted:

1. the location of each analyte line peak position relative to the Mn profile line.
2. the aspiration of high concentration single-element standard solutions and subsequent analysis using the spectrometer line array to determine potential interferences.
3. the re-scanning of each line during aspiration of a low concentration analyte standard solution, and multiple higher concentrations of the interfering element to determine the extent of the interference.

The evaluated ARL ICP has a mechanical spectral scanning device (this device is called a "SAMI" by the manufacturer) that drives the primary slit using motor steps in order to scan peak profiles over a specific wavelength range.

Scans were setup on the SAMI for ±80 motor steps to allow for sufficient resolution of the analyte peak regardless of spectral order. This "window" was sufficient to view any interferences close enough to the analyte peak which may be erroneously identified as the analyte. The scans were performed in 4-step increments and 2-s signal integration time was used at each step.

Two sets of scans were analyzed in order to determine spectral interferences.

a) First spectral interference scan
These scans were used as a screening technique to determine which elements would pose a potential interference for any given spectral line.

b) Second spectral interference scan
Information from these scans demonstrated the extent of each interference by showing the relative position of the analyte peak to the interferant peak and the ratio of the response of the interferant to the analyte standard for the particular spectral line.

After visually identifying potential spectral interferences from these peak profiles, the quantitative effects from these interferences were determined by measuring the intensities of each interfering element. Apparent concentrations resulting from these spectral line interferences on other channels were then determined. The IEC factor was determined for each interference using the following equation:

K=B/C

where:

K	=	IEC factor
B	=	Apparent concentration of the affected element
C	=	Concentration of the interferant (single-element) standard

Further details for determining spectral line interferences for the ARL 3560 ICP has been described in reference 11.7.

3.2 Results

3.2.1 Peak Profiles (Motor Steps Off-profile)

The peaks for the eight element line array(s) were scanned using the "SAMI" in order to determine how close they were to theoretical values. It has been suggested by the instrument manufacturer, as a general rule, that each spectral line should be within ±8 motor steps from the profile peak.

[Note: For this instrument, Mn and Cd are used as profile elements. Prior to each analysis, a solution containing the profile element(s) is scanned, the peak intensity identified, and the slit settings adjusted to give maximum intensity for the profile element(s). Further details regarding instrument profiling can be found in references 11.1 - 11.3, and 11.7.]

As summarized below, all the lines examined in this particular ICP-AES instrument were within ±2 motor steps.

Motor Steps "off profile"	Element
2	Be
0	Cd, Cu, Sb, Sn
-1	Ag, Pb
-2	Zn

This indicates that, as a general rule, when the instrument is profiled prior to each analysis, the line peaks of the solder elements are well within specifications.

3.2.2 Spectral Line Interferences

The interference relationships on specific wavelengths used for elements expected to be collected in solder/brazing operations are shown in Figures 1 and 2. Figure 1 is for those elements affecting Pb and Sb. Figure 2 is for those elements affecting Ag, Be, Cd, Cu, Sn, and Zn.

A chart summarizing significant spectral line interferences found during this evaluation is shown in Table 6. This Table lists the spectral line corrections in the order of correction (i.e. the interference of Cd on the Co channel is corrected before the interference of Co on the Cd or Pb channel).

As can be seen from the first correction in Table 6, an IEC correction value of 0.00021 is necessary to compensate for the affecting element Cd on the Co channel. If Cd is present, 0.00021 µg/mL is subtracted from any Co concentration for each 1 µg/mL of Cd present in the sample. If 10 µg/mL of Cd is present, then a 0.002 µg/mL signal is subtracted from the Co channel.

The range of IECs varied from a low of 0.000025 for Co interfering on the Cd channel to a high value of 0.0039 for Be interfering on the Sb line.

Because some solder samples may contain Bi, a study was conducted to determine the feasibility of analyzing and correcting for Bi. An interference from Bi on Sb and Pb was noted. The instrument as set up did not contain a channel for Bi. Bismuth is a component of some solders, especially the more recent "lead free" type. To compensate for this, the Se channel at 196.09 nm was modified to allow a scan of the Bi line at 196.006 nm using the SAMI scanning device. This modification to include the semiquantitative determination of Bi and any interferences from Bi on other lines requires additional analysis time because the SAMI offsets from the Se to the Bi wavelength for Bi measurements.

4.1 Procedure

After the IEC factors were entered into the ARL computer software, the precision and accuracy of the method were evaluated by analyzing spiked samples. The spiking scheme for the multi-element samples was conducted in the following manner:

The precision and accuracy for each element was evaluated by analyzing 18 spiked filter samples. Aqueous reference standards described in Section 1.3. were used for each spike. Spike amounts were calculated for levels at about 0.5, 1, and 2 times the OSHA TWA PEL assuming a 480-L air volume. A worst case scenario of air volumes < 400 L was assumed for samples containing about 0.5 times the PEL for silver. Each multi-element spike was delivered to a single mixed-cellulose ester (MCE) filter. The spikes were delivered from stock solutions using calibrated micropipettes. A calculation error occurred regarding the mass of silver during preparation of the first multi-element spikes. Filters spiked with silver at about 0.5 times the PEL were then prepared separately.

All spiked sample filters were digested and prepared for analysis using the procedure specified in the method (11.1). Each sample was diluted to a 25-mL solution volume. Samples separately prepared for silver (approximately 0.5 times the PEL) were diluted to a final volume of 10 mL. This solution volume is recommended in the method (11.1) to provide increased sensitivity for silver.

The instrument was calibrated as stated in Section 1.4. These samples were determined using a two-point calibration consisting of standards listed in Table 3 and a reagent blank.

4.2 Results

All sample results were examined in terms of precision (CV) and amount of error. Analytical error (AE) for each element is calculated as:

± AE % = 100 [|Mean Bias| + 2(CV)]      95% confidence

From the summary in Table 7, it can be seen that the AE ranged from 2.3% for antimony to approximately 17% for silver and lead. The precision range (CV) was from 0.01 for many of the analytes to 0.06 for silver and lead. The bias varied from -0.011 for beryllium to +0.055 for zinc.

5.1 Procedure

The procedure for the determination of detection limits, Background Equivalent Concentration (BEC) and short-term reproducibility for the ARL 3560 ICP has been previously described (11.7). The detection limits in this evaluation were determined after aspirating multielement calibration standards for a 5-s integration time. An 11 exposure sequence was used for the blanks and 10 exposures for the standards. Multi-element calibration standards (STND SOLNS 1 and 2) shown within Table 3 were used. The manufacturer's software algorithms calculate the qualitative detection limit (DL) as two times the standard deviation.

The qualitative detection limit for the ARL instrument is calculated as:

DL = (A × SDI × C)/(I - Io)

Where:

A	=	2 (qualitative detection limit) or
A	=	10 (quantitative detection limit)
SDI	=	Standard Deviation of the Background Intensity
I	=	Total Intensity
C	=	Analyte Concentration, the concentration of the calibration standard
Io	=	Background Intensity (determined from reagent blank)

The BEC is defined as the concentration of an analyte that is equal to the net intensity of the background signal for that analyte:
 

BEC = m(Io)

Where:

m = (C)/(I - I_o) = slope of the calibration curve
 

---

Note: The manufacturer's software algorithms automatically calculate the qualitative detection limit using two times the standard deviation (2SD) of the blank, among other considerations. Although the OSHA Inorganic Method Validation Protocol (11.9) states a qualitative limit shall be determined with 3SD_BL in the calculations 2SD_BL is accepted here to allow for future performance comparisons using the same instrument and software.

---

5.2 Results

The results for short-term instrument precision are reported in Table 8 as the Coefficient of Variation (CV) and Background Equivalent Concentration (BEC). At concentration levels equal to or greater than the BEC, short-term precision ranges are normally approximately 0.5% to 1.5% (as CV), provided linearity of the spectral response function is maintained. Short-term precision should be 1% or better for simultaneous systems (11.10). As can be seen in the last column of Table 8, the CV is < 1% for all lines. The qualitative detection limits varied from 0.001 µg/mL for beryllium to 0.05 µg/mL for lead. The BEC values ranged from 0.02 µg/mL for beryllium to 1.9 µg/mL for lead.

Detection limits determined approximately one year after these results were obtained show improved performance for most elements:

ELEMENT	QUAL DL
Ag Be Cd Cu Pb Sb Sn Zn	0.0043 0.0002 0.0054 0.0055 0.0425 0.0481 0.0396 0.0031

Detection limits are very dependent on maintenance and operating conditions, and should be periodically checked to assess instrument performance and the need for maintenance.

6.1. Procedure

The linearity was evaluated using a procedure previously described (11.7). The linearity of the calibration curves was checked by analyzing several standard solutions within the range from 5 to 10 times the calculated detection limit up to 1,000 µg/mL. The upper range was limited either by reaching a detector saturation level or by exceeding the value of the highest standard stock solution used (1,000 µg/mL).

6.2. Results

The highest concentration of most of the standards used was 1,000 µg/mL. For a few elements, a 1,000 µg/mL concentration did not saturate the detector nor was the linear range exceeded. Standard solutions exceeding a concentration of 1,000 µg/mL were available for these elements and linearity determinations beyond 1,000 µg/mL were performed. These exceptions were: aluminum, magnesium, lead, bismuth, and antimony.

The upper working ranges (µg/mL) and the concentrations at which the photomultiplier tubes (PMTs) become saturated for elements at the wavelengths designated in the array are summarized in Table 9. The PMTs should become saturated at a value between the Upper and Saturation concentrations reported in the Range column.

In Table 8, column five contains the lower quantitative detection (LQD) limit and column six the upper range of linearity for the brazing and solder elements. For the ARL instrument, the LQD is five times the qualitative detection limit. The quantitative detection limits ranged from 0.005 µg/mL for Be to 0.2 µg/mL for Pb and Sb.

The linear range was evaluated using a 5-s integration time. The upper concentration limit for the validated elements ranged from 20 µg/mL for Be to 1,000 µg/mL for Pb and Sb. The optimum working range for most elements exceeded 100 µg/mL.

7.1. Procedure

Sets of QCs were determined for a final evaluation of the precision and accuracy of the procedure. These QCs were independently prepared on spiked filters. The concentration levels for some of the analytes on these QCs were lower than those in the previous precision and accuracy evaluation. With the exception of beryllium, the concentrations for the prepared spikes were based upon calculations using a 4-h sampling period and an air volume of 480 L.

The concentrations ranged from 0.1 to 1 times the TWA PELs as follows:

ELEMENT	× PEL
Ag Be Cd Cu Pb Sb Sn Zn	0.5 (Ceiling or STEL) 0.1 0.5 to 1 1 0.5 to 1 0.5 to 1 1

To determine if precision and accuracy could be improved using smaller solution volumes, all samples were diluted to a final solution volume of 10 mL.

7.2. Results

The data, as summarized from Table 10, provide the following method performance information:

With the exception of Be in one QC set, the AE was less than 25%. The Precision (CV) was better than 0.14 for all samples. With the exception of Cu in the last QC set, the Mean varied within ±10%. For the most part, these results are similar or better than those presented in Section 4, and indicate a 10 mL solution volume can be used as an alternative dilution. It is unknown why one set of beryllium samples was ± 35% AE. Samples to be analyzed for beryllium at Ceiling or STEL levels should be diluted to 10 mL to improve sensitivity.

9.1. Procedure

Three "blind samples" containing the SRM were routinely analyzed by a laboratory chemist. Sample preparation for bulk material was carried out according to OSHA method no. ID-206. Additional details are described below:

sample wt range (mg)	sample wt median (mg)	final sample vol (mL)	sample (mg/mL)
10-14	12	250	0.048

9.2. Results

Due to the weights of SRM used, the values found for Ag were below both the quantitative and qualitative detection limits. The values for Cu, Sb, and Zn were below the quantitative but above the qualitative detection limits. Therefore, the values for Cu, Sb, and Zn could not be used and merely indicate that these elements were present in the sample matrix.

Although Zn was found in the SRM, it was not reported on the accompanying certificate. It is possible that due to the large amounts of Pb and Sn present in the bulk material these elements might possibly give an apparent value for Zn, as the spectral corrections for this method are for low concentrations typically found in workplace atmospheres. Caution must be observed when using this method as a screening tool to identify unknown elements in a bulk matrix. Minor elemental quantities found for elements in bulks must he considered as apparent values only, and may have to be confirmed by other methods such as Atomic Absorption Spectroscopy.

From the results reported below it is evident that if the minor components are to be reported, two sample aliquots will have to be taken; one for the more concentrated aliquot used for quantitating the values for Sb, Cu, and Ni, the other a larger weight for minor components.

A suggested scheme for solder bulk materials is to weigh a 10 to 20 mg aliquot; dilute to 250 mL to obtain the major components within the linear working range; and prepare another sample using approximately a 100-mg sample weight and 100-mL sample solution volume. This will enable both major and minor components to be identified, provided both sample results are carefully scrutinized for exceeding upper linear range limits, detection limits, or interferences potentially existing in a variable sample matrix.

	------------------Wt %--------------------
	Pb	Sn	Sb	Cu	Ni	Ag
Reference values	60.17	39.3	0.43	0.011	0.012	0.01
Avg Recovery	60.4	38.1	n.d.	n.d.	n.d.	n.d.
F/T	100.	96.9	-	-	-	-
n.d. = none detected

10.1. Introduction

An ideal situation for emission spectroscopy would be to have many alternate spectral lines available for analysis of each element to assist in characterizing the sample and minimize error. Unfortunately, there is a finite amount of physical space available for the installation of spectral lines in the array of a simultaneous direct reading emission spectrometer. This limits the number of element channels that may be installed. An alternative to increasing channels for the ARL 3560 ICP is to "create" lines using the SAMI scanning mechanism. The SAMI, as previously described in Section 3.1., is a stepper-motor-controller which allows for mechanical profiling of the slit image and the relative wavelength position. The SAMI is also routinely used for background corrections, moving off and on a fixed wavelength at specified intervals. This ability to move away from the fixed line position allows for examination of other emission lines nearby. These emission lines, referred to as "free SAMI" lines, are limited because they must be in close proximity to element channels previously installed by the instrument manufacturer.

10.2. Procedure

Using the scanning ability of the SAMI mechanism, free SAMI lines (optional wavelengths) for the elements Bi, Cd, Be, and As were programmed into the manufacturer's software as described in the table below:

Selected Free SAMI Lines

Ele(2)	Wavelength(2) (nm)	Ele(1)	Wavelength(1) (nm)	Offset (Steps)	Spectral Order
Bi Cd1 Cd3 Be3 As1	196.006 228.801 214.433 217.506 228.812	Se Co Te Sb Co	196.090 228.616 214.275 217.581 228.616	218.00 -320.00 -274.00 192.00 -338.00	3 2 2 3 2
Where: Ele(1) = original channel modified to allow a scan of Ele(2). For example, the Co 228.616 nm line [Ele(1)] was modified by the SAMI (offset -320.00 divisions) to scan the 228.801 nm line for Cd [Ele(2)].

Detection limits and BECs were then determined for these free SAMI lines using the procedure discussed in Section 5.

10.3. Results

The following BEC and detection limit (DL) values were calculated after determinations using the concentration of standards specified (Std Concn). Reagent blanks were prepared in DI H₂O. The DL was reported as a qualitative detection limit (A=2, as stated in Section 5.1).

BEC/DL - Free SAMI and Original Spectral Lines
 

Channel Name	Blank Int.	Blank SD	Std Int.	Std Concn	BEC	DL	Wavelength (nm)
As	5.8799	0.2734	143.98	100.00	4.2578	0.3960	189.042
As1	6.1419	0.1108	67.950	100.00	9.9371	0.3586	228.812
As	2.7624	0.1230	1,369.1	1,000.0	2.0217	0.1801
As1	4.7699	0.0612	625.87	1,000.0	7.6798	0.1971
Be	4.5310	0.2074	1,450.4	10.000	0.0313	0.0029	313.042
Be3	4.3889	0.0232	29.060	10.000	1.7790	0.0188	217.506
Be	4.5310	0.2074		100.00	saturated detector
Be3	4.3889	0.0232	246.92	100.00	1.8096	0.0191
Bi	3.3873	0.0172	7.1453	100.00	90.134	0.9166	196.006
Bi	3.3527	0.0302	11.011	200.00	87.555	1.5754
Cd1	4.2365	0.0168	203.77	10.000	0.2123	0.0017	228.801
Cd3	3.5647	0.0164	314.33	10.000	0.1147	0.0011	214.433
Cd	1.7259	0.0164	98.449	10.000	0.1784	0.0034	226.502
Cd1	4.2365	0.0168	1,000.0		saturated detector
Cd3	3.5647	0.0164		1,000.0	saturated detector
Cd	1.7259	0.0164		1,000.0	saturated detector
The elements (Channel Name) which are not followed by a number are the original installed lines. Blank Int. and Std Int. are blank and standard intensities in K-pulses (this is a unique term used by the manufacturer to designate intensity counts). All concentration (Concn) values (Std, Concn, BEC, and DL) are in µg/mL.

These additional lines may be used as elemental confirmation to avoid the reporting of "false positives" which can occur at concentrations near the detection limit due to matrix effects, uncharacterized spectral, or other interferences. The application of these alternate lines will extend the analytical time taken for each sample, and subsequently affect the amount of sample volume used during sample preparation. Sample solution volumes of 10-mL may be insufficient if numerous additional lines are scanned and more than one sample determination is necessary. Some notes regarding the sensitivity of each line are listed below:

Bismuth:
The Bi channel is relatively insensitive, yielding high BEC and DL values at concentrations of from 100 to 200 µg/mL. However, as previously discussed in Section 3, this line may be used as confirmation for the presence of Bi in "lead free" bulk solder samples.

Cadmium:
The alternative lines for Cd have improved detection limits of a magnitude of about three times that of the Cd 226.502 nm line. This increased sensitivity is important as the PEL for Cd has recently been lowered significantly (29 CFR 1910.1027 or 29 CFR 1926.63). It is important to note there is a major As spectral interference on the alternate Cd 228.801 nm line.

Arsenic:
Although at first glance it appears that the alternate As 228.812 line has a comparable detection limit with the As 189.042 line, this is misleading. Upon further examination, intensities from the same sample concentrations are about two times greater for the As 189.042 line. The reason for the similar detection limits results from the imprecision of the As 189.042 blank (being about two times higher than that of the As 228.512 line). In addition, there is a major spectral interference on the alternate As line resulting from cadmium.

Beryllium:
The original Be 313.042 line is approximately ten times more sensitive than the alternate Be 217.506 line; however, the 217.506 line has a higher upper range and reduces the requirement of further sample dilution for samples containing high concentrations of Be.

Table 4
Air Contaminants - OSHA Permissible Exposure Limits¹

Element	Substance Exposed to	Transitional PEL (mg/m³)		Final Rule PEL (mg/m³)
		TWA	CEILING	TWA	STEL	CEILING
Ag	Metal dust and fume (as Ag)	0.01		0.01
Be2	Be and compounds (as Be)	0.002	0.005³	0.002	0.005³	0.025
Cd	Fume Dust	0.1 0.2	0.3 0.6	0.1 0.2	0.3 0.6
Cu	Fume (as Cu) Dusts amd mists (as Cu)	0.1 1		0.1 1
Pb	Inorganic	(see 29 CFR 1910.1025, PEL is 0.05 mg/m³)
Sb	Sb and compounds (as Sb)	0.5			0.5
Sn	Inorganic compounds except oxides (as Sn) Tin oxide (as Sn)	2		2 2
Zn	Zinc chloride fume Zinc oxide fume	1 5		1 5	2 10
Note: underlined values are those TWA PELs selected for validation.
STEL duration is for 15 min unless otherwise noted. 1 From reference 11.4 2 Beryllium also has Transitional Peak PEL concentration limit of 0.025 mg/m³. 3 Both the Transitional Ceiling Limit and the STEL for beryllium are    for a maximum 30-min duration.
Note: This validation was performed using the Final Rule Limits bolded above. With a couple of exceptions the Final Rule Limits for the elements or compounds listed are identical to the Transitional Limits. Analysis of samples by the ARL 3560 using OSHA method no. ID-206 should perform well regardless of whether Transitional or Final Rule Limits are used. After this validation was performed, the Final Rule Limits were vacated. For determination of exposures to rosin core solder pyrolysis products, please see OSHA method no. 54 for formaldehyde.

Table 5
Common Solder Alloys*

---------Composition %---------						Melting range, °C	Use
Sn	Pb	Cd	Bi	Ag	Sb
63	37	-	-	-	-	183	eutectic solder for electronic application
60	40	-	-	-	-	183-190	high quality solder
50	50	-	-	-	-	183-216	general-purpose solder, plumbing
40	60	-	-	-	-	183-238	wiping solder, radiator solder
30	70	-	-	-	-	183-255	machine and torch soldering
20	80	-	-	-	-	183-277	automotive-body solder
95	-	-	-	-	5	235-240	refrigeration soldering
x	0	0	x	x	0		"Pb free" plumbing solderinga
62	36	-	-	2	-	179	soldering silver surfaces
1	97.5	-	-	1.5	-	309	high temperature soldering
15.5	32	-	52.5	-	-	90	fusible links
13	27	10	50	-	-	70	low melting solder
* Modified from reference 11.5 - not present or reported a This solder also contains copper, and is patented alloy containing unspecified amounts of elements designated with an x. These were confirmed at the SLTC by ICP and XRF analysis. A sample of "Safe Flo (TM) Silver" solder, gauge 0.084, from Oatey, Cleveland, OH 44135 was used for this confirmation. Note: Because of recent requirements of the Safe Drinking Water Act and certain state laws restricting lead-solder has been developed that contains no lead, antimony, or cadmium. These "lead free" solders have become popular in the marketplace.

Table 6
Interferences

	Channel	Int.	IEC factor
	Co V Cd Cd Sn Sn Sn Cu Ag Ag Pb Pb Pb Pb Sb Sb Sb Sb Zn Be Sb Sb Zn Ag Sn Sb Sb Pb Bi* Bi* Bi*	Cd Be Ni Co Al V Mo Mo Mn V Ni Co Al Mo Ni Mo Fe1 Al V Ni V Zn Pb Cu Cu Be Be Bi* Bi* Be Fe Al	0.00021 0.00093 0.00011 0.000025 0.00040 0.0016 0.00013 0.00061 0.00012 0.00016 0.00074 0.00044 0.0010 0.000235 0.0015 0.00035 0.00022 0.0012 0.0014 0.0038 0.00046 0.0015 0.00081 0.0013 0.000062 0.00042 0.0039 0.0003 0.001 0.009 0.37 0.049
Int. Channel = Interference = The Affected Channel * Bi line available through modified Se channel.

This table is organized in the following manner:

The first two columns list two elements. The first element is the channel that is affected by the second element (the affecting element). The inter-element Correction Factor (IEC) for the affected element is shown in the third column.

Table 7
Precision and Accuracy

	Element
	Ag	Range taken (µg) Bias CV1 (Pooled) Analytical Error (Total)	= = = =	2.21 to 10.1 +0.036   0.066 ±16.9%
	Be	Range taken (µg) Bias CV1 (Pooled) Analytical Error (Total)	= = = =	0.5 to 2.0 -0.011   0.015 ±4.0%
	Cd	Range taken (µg) Bias CV1 (Pooled) Analytical Error (Total)	= = = =	24.0 to 100. +0.054   0.026 ±10.5%
	Cu	Range taken (µg) Bias CV1 (Pooled) Analytical Error (Total)	= = = =	24.0 to 100. +0.017   0.039 ±9.5%
	Pb	Range taken (µg) Bias CV1 (Pooled) Analytical Error (Total)	= = = =	12.0 to 48.0 +0.050   0.058 ±16.7%
	Sb	Range taken (µg) Bias CV1 (Pooled) Analytical Error (Total)	= = = =	120. to 480. +0.008   0.007 ±2.3%
	Sn	Range taken (µg) Bias CV1 (Pooled) Analytical Error (Total)	= = = =	480. to 1,920. +0.033   0.015 ±6.3%
	Zn	Range taken (µg) Bias CV1 (Pooled) Analytical Error (Total)	= = = =	600. to 4,800. +0.055   0.013 ±8.0%
Note: Analytical errors (AE) of < ±25% are acceptable for all elements validated except lead. [Lead is required to have < ± 20%, see 29 CFR 1910.1025(d)(9) for further information]

Table 8
Detection Limits, BEC, Ranges, and Short-Term Precision

ELEMENT	WAVELENGTH		*QUAL	----*RANGE----
	(nm)	*BEC	DL	LOWER	UPPER	CV
Ag Bi Be Cd Cu Pb Sb Sn Zn	328.068 196.006 313.042 226.502 324.754 220.353 217.581 189.980 213.856	0.4857 90. 0.0177 0.1645 0.4982 1.8816 1.5882 0.6104 0.1897	0.0109 1.0 0.0010 0.0131 0.0142 0.0470 0.0388 0.0139 0.0241	0.0545 5.0 0.0050 0.0655 0.071 0.235 0.194 0.0695 0.1205	500 1,000 20 250 250 1,000 1,000 900 500	0.0035 0.03 0.0045 0.0031 0.0019 0.0066 0.0092 0.0015 0.0028
Results were obtained from determinations performed on 21-Jan-92 * = BEC, DL, and RANGE values reported as µg/mL DL = Qualitative Detection Limit (SD = 2) BEC = Background Equivalent Concentration CV = Short-Term Precision (based on three exposures) LOWER = Quantitative Detection Limit (5 × DL) UPPER = Upper Working Range Integ. time = 5 s

Table 9
Upper Range and Saturation Concentrations

			-----Range-----
Element Name	Wavelength (nm)	Order	Upper >		Saturation <
Ag* Al As Be* Bi Ca Cd* Co Cr Cu* Fe1 Fe2 Mg Mn Ni Pb* Sb* Se Si Sn* V Zn*	328.068 308.215 189.042 313.042 196.006 393.366 226.502 228.616 267.720 324.754 259.940 271.440 279.080 257.610 231.604 220.353 217.581 196.090 288.158 189.980 310.230 213.856	2 2 3 2 3 2 3 2 3 2 3 3 2 3 3 3 3 3 2 3 2 3	500 5,000 700 20 10,000 15 250 300 300 250 200 400 3,000 60 600 1,000 1,000 1,000 1,000 900 1,000 500		1,000 10,000 50 10,000 500 500 500 500 5,000 100 1,000 3,000 3,000 1,000 1,000 1,000
Range is listed as µg/mL * Validated elements

Table 10
Quality Control Samples

	N	MEAN	STD DEV	CV	AE (±%)		µg TAKEN ---(range)---
QC Set 1
Ag Be Cd Cu Pb Sb Sn Zn	3 3 3 3 3 3 3 3	0.930 0.962 0.935 1.089 0.939 0.898 0.917 0.937	0.023 0.015 0.021 0.027 0.024 0.047 0.023 0.011	0.025 0.015 0.023 0.025 0.026 0.052 0.025 0.012	11.9 6.9 11.1 14.0 11.2 20.6 13.3 8.7		2.41 0.201 5.02 25.2 20.1 126. 504. 315.	3.66 0.305 7.62 47.4 30.5 237.0 948. 592.
QC Set 2
Ag Be Cd Cu Pb Sb Sn Zn	3 3 3 3 3 3 3 3	0.998 1.091 0.998 0.993 1.023 0.933 0.964 0.976	0.013 0.141 0.010 0.011 0.021 0.023 0.012 0.007	0.013 0.130 0.010 0.011 0.020 0.024 0.013 0.007	2.9 35.0 2.3 3.0 6.4 11.5 6.1 3.9		2.77 0.228 5.77 33. 23.1 165. 660. 412.5	3.46 0.267 7.20 51. 28.8 258. 1032. 645.
QC Set 3
Ag Be Cd Cu Pb Sb Sn Zn	3 3 3 3 3 3 3 3	0.927 0.934 0.898 1.139 0.930 0.901 0.894 0.912	0.005 0.039 0.043 0.029 0.035 0.025 0.023 0.029	0.006 0.041 0.048 0.026 0.038 0.027 0.025 0.032	8.5 14.9 19.8 19.0 14.6 15.3 15.7 15.2		2.60 0.217 5.42 34.8 21.7 174. 696. 435.	3.43 0.286 7.15 54.0 28.6 270. 1080. 675.
AE = Analytical Error

 

SPECTRAL LINE INTERFERENCES Figure 1: Interactions of Pb and Sb

SPECTRAL LINE INTERFERENCES Figure 2: Interactions of Ag, Be, Cd, Cu, Sn, Bi and Zn

i.e. as shown from Figure 2, Be has an interference from V; Be interferes with V, Sn, Bi and Al.