TTP No.: CH131001
Revision: 2
Date: 12/6/93
Old TTP No.: None
Title: INFRARED ANALYSIS OF WASTES: NOVEL LABORATORY AND ON-LINE MEASUREMENTS
BY PHOTOACOUSTIC AND TRANSIENT INFRARED SPECTROSCOPIES
Contractor: Ames Laboratory/Iowa State University
Responsible Division: R&D
FY: 1994
Technical Program Officer:
Steve Webster, DOE-CH, 708-252-2822
Principal Investigators:
John McClelland, Ames Laboratory, 515-294-7948
Roger Jones, Ames Laboratory, 515-294-3894
Technical Program Manager:
James Corones, Ames Laboratory, 515-294-9636
Joint Participants:
Westinghouse Hanford (Teofila Rebagay, 509-373-3174)
Rocky Flats Plant (Andrea Faucette, 303-966-6420)
Brookhaven National Laboratory (Paul Kalb, 516-282-7644, TTP No. CH321202)
Jointly Funded Program Code:
Integrated Program: IP2 CHARACTERIZATION, MONITORING, AND SENSOR TECHNOLOGIES
Primary Technology Areas:
WBS Element: TBD
B&R Code:EW40
SUMMARY
Task Summary:
Molecular analysis of inorganic wastes in unprocessed sludge and processed waste forms is vital to the success of DOE clean-up efforts. In Subtask 1, the project will develop the methodology for the quantitative analysis of molecular and multi-atom ionic species in the tank sludges and solids at Hanford via Fourier-transform infrared (FTIR) photoacoustic spectroscopy (PAS). Such species are of special interest because of their chemical reactivity. For example, ferrocyanide, nitrate and nitrite ions potentially can react explosively. FTIR-PAS can identify and quantify these species in the sludges and solids.
PAS is a technique that allows the power of optical spectroscopy to be used on materials whose physical characteristics normally prevent such an approach or demand ex tensive sample preparation. For quantitative PAS, all that is required is that a known or reproducible amount of sample be placed in a PAS detector cell. There is no other sample preparation. In addition, FTIR-PAS reduces the amount of sample required compared to conventional techniques, reducing radiation hazards accordingly. The minimum sample size in the FTIR-PAS analysis of sludges is typically on the milligram level. The general scheme for the sludge analysis is shown in the figure. FTIR-PAS samples can be small enough to be removed from the hot cell or other containment. They are then simply sealed in the PAS cell and the analysis proceeds without further handling.
he methodology for rapidly and routinely producing quantitative FTIR-PAS spectra from small surrogate-waste samples with a minimum of sample handling is being worked out in Subtask 1. Another activity, which will proceed during the whole of the project, is the analysis of exactly how the FTIR-PAS spectra depend on the complex composition of the waste materials. Once the methodology for producing quantitative spectra has been worked out, a training session will be held at Hanford by Ames personnel. The FTIR-PAS instrumentation, the quantitative methodology, and as much of the information on spectral dependence on composition that is determined will be transferred to Hanford personnel at that time. The spectral dependence on composition will then continue to be studied, and the results transferred to Hanford as they become available. An automatic sampling and shuttle system for use in hot areas will also be developed and installed at Hanford. Subtask 2 addresses the problem of process safety and control and waste certification in the polyethylene encapsulation of low-level radioactive waste, which is being developed at Brookhaven and implemented at Rocky Flats and Hanford. In polymer encapsulation the salt waste is mixed, heated, and extruded with a synthetic polymer. The goal of this subtask is the development of a system to monitor in real time the salt-to-polymer ratio of the molten, extruded waste stream. Real-time measurement of this ratio will allow operators to control the process so that the final waste form has the optimum composition. Too high of a ratio results in poorly encapsulated, inadequately immobilized waste. Too low of a ratio is inefficient; too little waste is encapsulated in too large a volume. The real-time monitor will also insure uniformity in the encapsulated waste by alerting operators to fluctuations in the salt-to-polymer ratio that signal poor encapsulation of a portion of a batch, even when the average ratio of the batch is correct. Ultimately, the data from the monitoring system can be used to document the waste composition for purposes of certifying the final waste form.
The monitoring system will use the new technique called Transient Infrared Spectroscopy (TIRS), which can record in real-time the mid-infrared spectrum of a moving solid or viscous-liquid stream. The TIRS technology has previously been developed as an on-line compositional and physico-chemical analysis system for monitoring product and feedstock streams on industrial production lines. It has been successfully tested on the process lines of several manufacturers and was selected for a 1992 R&D 100 Award. The figure shows how TIRS works. A jet of gas produces a hot or cold surface layer on the flowing stream. This layer is then observed by an infrared spectrometer, so it is the infrared spectrum of the layer that is acquired and used for the analysis.
Successful laboratory-scale proof-of-principle tests using loaned equipment on an extruder at Brookhaven National Laboratory have been completed. An FTIR spectrometer for the monitor and a laboratory-scale extruder are being acquired, and they will be used to produce a TIRS monitor optimized for the salt/polymer waste monitoring application. It will then be demonstrated both on equipment and at sites chosen by staff at the user sites. After that, the monitor will be scaled up and improved for a test on the pilot-scale waste-processing system. This pilot-scale demonstration is planned for early 1994. At the end of the pilot-scale test, another key milestone will have been reached. The decision will be made as to which of two TIRS versions (heating or cooling) will continue to be developed. A simplified, lower-cost, filter-based TIRS unit will then be custom designed for the specific needs of the waste monitoring application.
TASK/MILESTONE SUMMARY
(Click here for time chart - see below for numbers)
Project Tasks/Milestones:
9.-11. Document submittal. 9/93-11/93.
Subtask 1
1. Acquisition of FTIR-PAS systems. 12/92 - 9/93.
2. Correlate spectra with composition using surrogates. 12/92 - 3/95.
3. Establish methods for acquiring quantitative spectra. 12/92 - 11/93.
12. FTIR-PAS installation and training session at Hanford. 1/94. [*¶; A]
13. Continue transference of expertise. 2/94 - 9/95.
14 & 19. Develop and transfer shuttle. 5/94 - 8/95. [¶; D]
Subtask 2
4. Proof-of-principle test. 12/92. [§]
5. Procure and prepare instrumentation. 1/93 - 8/93.
6. & 8. Develop lab-scale TIRS and components for pilot scale tests. 3/93 - 1/94.
7. Develop preprocessing analysis. 4/93 - 4/94.
15. & 16. Install and test system on pilot-scale line. 1/94 - 3/94. [*¶; C]
17. Develop chosen version of TIRS. 4/94 - 7/95.
18 & 20. Design, build, and deliver filter-based TIRS unit. 5/94 - 9/95. [¶; E]
21. Write R&D Status Report. 8/95 - 9/95.
Code: § = Completed Milestone; * = Key Milestone; ¶ = Planned
Milestone
Milestone Explanations:
A. Key Milestone: Determine that sufficient methodology and expertise for experimentation to begin have been transferred to Hanford.
B. Key Milestone: Decide whether application of TIRS technology should be developed.
C. Key Milestone: Choose between heating and cooling versions of TIRS for further development.
D. Install sampling and shuttle system at Hanford.
E. Deliver filter-based TIRS unit to user site and train personnel there.
BUDGET SUMMARY AND DATA
BUDGET SUMMARY (in thousands of dollars):
FY93-ACT FY93-CO FY94-ACT FY95-R
OP 200 23 323* 300
EQ 100 0 0 0
GPP 0 0 0 0
LI 0 0 0 0
Total 300 23 323* 300
* Includes $23K carry over from FY93.
TECHNICAL DESCRIPTION
Technical Task Description
Subtask 1
The waste tanks at Hanford Site represent one of the most demanding of waste management problems in the DOE complex. The contents of the tanks are chemically and physically highly variable and are chemically active. A thorough chemical analysis of the contents is therefore necessary before they can be safely handled, much less disposed of. The sludges and solids that have formed in the tanks are particularly demanding analytically. Their physical characteristics require that samples must be extensively prepared before they can be submitted to the conventional instrumental techniques available in the modern analytical laboratory. On the other hand, the high radioactivity of the sludges and solids demands that handling be kept to a minimum. This project proposes part of the solution to this conundrum.
Subtask 1 will develop the methodology for the quantitative analysis of molecular and multi-atom ionic species in the sludges and solids via Fourier-transform infrared (FTIR) photoacoustic spectroscopy (PAS). PAS is a technique that allows the power of optical spectroscopy to be used on materials that are opaque, rough, and difficult to handle. PAS requires only that a small, known quantity of the sample material be sealed in a photoacoustic detector cell before it can be analyzed, thus handling is kept to a minimum. In addition, the minimum sample required for FTIR-PAS of solids and pastes is small, typically in the milligram range. This reduces the radiation hazard and opens the possibility that much of the analytical work can be done outside of a hot cell, considerably increasing the efficiency of the process. Proof-of-principle experiments on surrogate sludge material provided by Hanford Site have already been successfully carried out, as described in the Background section below. It appears that important questions about the sludges can be answered. For example, there is a special concern over the amounts of nickel ferrocyanide ([NiFe(CN)6]²¯, nitrate ([NO3]¯), and nitrite ([NO2]¯) ions present in the tanks because of the following potentially explosive reactions:
[NiFe(CN)6]²¯ + 6[NO3]¯ -> 6N2 + 2CO2 + 4[CO3]²¯ + FeO + NiO
[NiFe(CN)6]²¯ + 10NO2 ¯ -> 8N2 + 6CO3²¯ + FeO + NiO
Strong features arising from [NiFe(CN)6]²¯ and [NO3]¯ were observed in the FTIR-PAS spectra of these surrogates. These features should permit a quantitative analysis for these species. There was no [NO2]¯ in the samples provided.
At present there are two main activities in the subtask at Ames Laboratory. First, the analysis of how the FTIR-PAS spectra vary as a function of the complex composition of the wastes has been started using surrogates. Second, the methodology for reliably acquiring quantitative spectra of milligram waste samples with minimal handling is being developed using surrogates. In January 1994, after the methodology for reliable spectra has been worked out, a training session will be held at Hanford by Ames personnel to transfer the PAS technology, the methodology for reliable spectra, and that portion of the spectral dependence on composition that has been worked out. Work at Ames will then continue on the spectral dependence on composition, still using surrogates, and that information will be transferred to Hanford as it is developed. In addition, work on an automatic sampling and sample shuttle system compatible with robotic operations will commence. This system will be designed for rapidly taking samples within a hot cell or other isolated environment for FTIR-PAS analysis.
Subtask 2
The polymer encapsulation of radioactive salt wastes is a promising new method for producing a certifiable, reduced-volume waste form that is being developed and demonstrated at Brookhaven National Laboratory, Rocky Flats Plant and Westinghouse Hanford. In the encapsulation process, the salt waste is mixed, heated, and extruded with a synthetic polymer. The goal of this subtask is the development of a system to monitor in real time the salt-to-polymer ratio of the molten, extruded waste stream. This ratio is important because too high of a ratio results in poor encapsulation and the wastes are not immobilized, while too low of a ratio is inefficient, resulting in a product containing less salt waste per volume than possible. The monitor will provide real-time data that will both guide the waste-processing operators in maintaining the optimum salt-to-polymer ratio and document the composition of the final waste form.
The monitoring system will use Transient Infrared Spectroscopy (TIRS). TIRS captures the infrared spectrum of a solid or viscous-liquid stream in real time and so can be used to monitor any property of the material stream that can be correlated with the infrared spectrum. In this subtask, the strengths of the infrared bands arising from multi-atom ions, such as nitrate, and from the polymer will be used to monitor the salt-to-polymer ratio. The TIRS technology was invented several years ago by the principal investigators, and in general TIRS is already well developed. It has been successfully demonstrated on the process lines of several manufacturers. Ames Laboratory recently received a 1992 R&D 100 Award when TIRS was chosen as one of the top 100 most technologically significant products, materials, processes, or software of 1991.
Successful proof-of-principle experiments involving a molten stream of polyethylene encapsulated salt were carried out at Brookhaven National Laboratory in early FY93. Some results from these experiments are illustrated in the following Background section. A decision (key milestone) was made at that time to proceed with the development of this application of TIRS. The equipment necessary for carrying out experiments at Ames Laboratory is being acquired and a laboratory-scale monitoring system will be developed. The system will be improved and then brought up to pilot scale in 1993. The improved system will be demonstrated in 1994 with extruders or other waste encapsulation machinery selected by staff from the user sites at locations arranged by the user-site personnel. In early 1994 the scaled-up version will be included in the pilot-scale technology demonstration being planned for Brookhaven (TTP No. CH321202). The results from this development and demonstration will be used to decide which version of TIRS (see Background for details) to continue developing for the application. The chosen version will then continue to be improved and hot tests at a user site will be included as needed. A simplified, lower cost type of TIRS monitor will be designed specifically for the waste-stream monitoring task. This custom monitor will be delivered to a user site at the end of FY95.
BACKGROUND
Infrared spectroscopy is ideal for measuring the composition of a material since the bands in the infrared spectrum of a material are directly related to the kind and quantity of chemical species present. It has long been a powerful tool in the laboratory for determining the composition of both simple and complex samples of all types. Conventionally, infrared light is passed through a sample to measure how much absorption occurs at different wavelengths (i.e., to measure the infrared spectrum). Solids and other opaque materials can be analyzed by this conventional transmission approach only after careful physical thinning or dissolving. This is not acceptable with materials like the tank sludges and solids, for which handling must be kept to a minimum, or for the on-line monitoring of the polymer-salt waste form, in which the analysis is required in near real time.
Subtask 1
The need for substantial sample preparation in conventional transmission spectroscopy has led to the development of newer, alternative methods like diffuse reflectance, attenuated total reflectance and photoacoustic spectroscopy (PAS) for spectroscopic analysis. Diffuse reflectance is best suited for materials in which light penetrates well into a sample and can be scattered from many surfaces, as is the case with powders and suspensions that are not strongly absorbing or can be diluted, so diffuse reflectance is not well suited to the present problem. Attenuated total reflectance involves spreading the sample on a transparent crystal and making intimate contact between the sample and the crystal. Light is then reflected off the sample-crystal interface. This technique could work with sludges of the correct consistency, but the sample must generally be a gram or more in size and the required crystal is not robust, either physically or chemically. It will be attacked by the chemically active and abrasive materials in the wastes, destroying the calibration of the instrument.
PAS is the most versatile of the techniques for analyzing intractable materials. It places the fewest limitations on the sample, which may be solid, liquid, or gas; transparent, milky or opaque; powdery or monolithic; microscopic or macroscopic. Handling can be kept to a minimum since the sample normally does not require any preparation. The sample merely needs to be sealed inside a PAS cell for the analysis. The figure at the top of the next page shows schematically how PAS for a solid sample works. A beam of intensity-modulated infrared light is sent into a sealed chamber containing the solid sample. Since most solids are opaque in the infrared, the light does not pass through the sample, as is required in conventional transmission spectroscopy. Instead, the light is absorbed within a distance 1/a of the sample surface, where a is the absorption coefficient of the sample. The absorption of the modulated light causes a modulated heating of the sample. The deposited heat diffuses to the sample surface and transfers to the surrounding gas. This causes a modulated heating of the gas, which also modulates the pressure of the gas inside the sealed cell. A microphone inside the chamber picks up these pressure fluctuations as sound waves, producing the PAS signal. Although this process may seem roundabout, it has a special advantage. Roughly speaking, only that portion of the heat deposited within a distance L of the sample surface by a light pulse can diffuse out of the sample before the next light pulse and contribute to the signal. L depends on the frequency of the light modulation, so the user can adjust it by changing the frequency. This lets the user adjust conditions so that good quality spectra can be acquired from materials that are very dark (high a) or nearly transparent (low a). No other technique has this adjustability.
Proof-of-principle tests have already been run, producing good results. Surrogate sludge samples provided by Westinghouse Hanford were examined by FTIR-PAS, and the spectra for 1 mg samples of two of them are shown in the upper figure on the right. For comparison, the lower figure on the right shows FTIR-PAS spectra of some of the pure components of the sludges. Some aspects of the compositional analysis are easy to see. For example, several of the sludge bands arise from the [NO3]¯ ion illustrated here by NaNO3 (bands at 2850, 2760, 2430, 1790, 1400, and 837 cm¯¹). Ferrocyanide is present in the sludges almost exclusively as the nickel ferrocyanide ion [NiFe(CN)6]²¯, as shown by the bands at 2100, 1620, and 1120 cm¯¹. Lastly, the two sludges differ by the amount of phosphate [PO4]³¯ present, as indicated by the band at 1040 cm¯¹, which is strong in the upper sludge spectrum but weak or absent in the lower one.
Subtask 2
In recent years manufacturers have increasingly used on-line analysis both to guide the real-time control of their process lines and to document the quality of the material produced. Laboratory and other off-line measurements are generally too slow to guide the control of a process line. On-line analysis is essential whenever the production of out-of-specification material can not be tolerated. Polymer-encapsulated waste is one such case; its quality must be documented and the production of out-of-tolerance material should be avoided because of the expense and hazards attached to reprocessing radioactive materials. The present measurement method used by those developing the encapsulation process is off line. The density of the final waste form is measured, and the ratio of salt to polymer is calculated from the known densities of the pure salt and polymer. Although this is sufficient during the development process, it has several shortcomings for the final production process. It is not real time; the ratio is determined only after the processing is complete, so the information can not be used to guide the processing. It measures the average of the complete waste form and so misses any fluctuations in the salt/polymer ratio that might signal poor encapsulation in part of the product. It assumes the density of the salt is invariant. It is unlikely that the salt waste will be that well characterized during the actual production process. Its density may vary substantially. The monitoring system proposed here will circumvent all of these problems.
Infrared spectroscopy is a powerful tool for chemical analysis, which potentially could be used on line. Gases are sufficiently transparent and low-viscosity liquids can be pumped through a thin enough spectroscopic cell to allow the use of conventional transmission spectroscopy on line. Solids and viscous liquids are another matter. Their infrared spectra can not be observed on line by conventional means because they are too opaque without being thinned and can not be squeezed through a sufficiently thin cell on line. Until recently, the mid-infrared spectra of such materials could only be measured in the laboratory after carefully thinning or diluting a sample.
Emission spectroscopy, in which a material emits a spectrum by virtue of its temperature, might seem to be a solution to this problem, since light is not required to pass all the way through a sample, but it too suffers from the opacity problem. When a thin body is heated, it emits an infrared spectrum that contains the same bands, the same information, as a conventional spectrum taken by passing infrared light through it; however, when a thick body is heated, only a featureless, ``blackbody'' spectrum is emitted. This is because an opaque body strongly absorbs the same bands it strongly emits when heated. This self-absorption smears out the spectrum bands and only the blackbody spectrum remains.
TIRS is a new approach that allows the mid-infrared spectrum of a stream of solid or viscous liquid to be determined as it moves along a process line. TIRS uses a temperature gradient to spectroscopically isolate a surface layer of the material. The layer is thin, so its spectrum has the bands needed for a compositional analysis; it is not smeared out. The figure at the top of the next page schematically shows how TIRS works. As the process material moves past the field of view of an infrared spectrometer, its surface is struck by a jet of either hot or cold gas, heating or cooling the surface of the material. This hot or cold surface layer will thicken by thermal diffusion, but since the material is in motion, the layer is carried out of the field of view as it thickens. In the field of view the hot or cold layer remains thin. If it is a hot layer, it acts as an emission source independent of the rest of the material stream, and the spectrometer records its emission spectrum. Since the layer is thin, its emission spectrum is not blackbody-like but structured. If the layer is cooled, it does not emit, but it does absorb infrared light passing through it. Blackbody like emission from the uncooled bulk of the material stream passes through the cooled surface layer on its way to the spectrometer. The spectrometer observes the blackbody envelope from the bulk of the material with the structured transmission spectrum of the cooled layer superimposed on it. The principal investigators have published the technical details of TIRS (see articles 4 and 5 listed in the References section) and have illustrated some of the potential industrial applications (see references 6, 7 and 8). Results from the proof-of-principle tests for this subtask (Milestone 5) have also been reported in the scientific literature (reference 8).
Proof-of-principle tests of a TIRS monitor for a molten salt/polyethylene stream have already been carried out. Both cold-jet and hot-jet versions of TIRS were tested at Brookhaven National Lab on an extruded stream of sodium nitrate (a surrogate for rad-waste salt) and polyethylene. The figure at the right shows the correlation observed during these tests between cold-jet TIRS results and the known actual sodium nitrate loadings in the stream. The correlation is very good, with a root-mean-square error of 3.5% by weight. The hot-jet version of TIRS produced results of similar quality, and it is not apparent at this time which version of TIRS will prove superior for this application.
TECHNICAL PROGRESS/MILESTONES
The timing of many of the following tasks and milestones must be considered tentative, since many of them are dependent on activities at the various joint-participant sites. Monthly expenditures of operating funds are expected to be approximately equal throughout FY94.
FY93
Subtask 1
1. Coordinate procurement of the equipment for two FTIR-PAS systems to insure transferability of data and methodologies to Hanford. Performance Period: December 1992 - September 1993. Completed.
2. Analyze FTIR-PAS spectra of surrogate wastes for qualitative speciation and quantitative correlation with composition. Performance Period: December 1992 - March 1995.
3. Establish methodology for procuring high quality, reproducible FTIR-PAS spectra using surrogate waste material at Ames Laboratory. Performance Period: December 1992 - March 1994.
Subtask 2
4. Prepare a first-prototype TIRS system using loaned instrumentation and run proof-of-principle tests of both versions (heating and cooling) of TIRS with it using the lab-scale extrusion system at Brookhaven National Laboratory. Performance Period: December 1992. Completed.
Milestone A. (Key Milestone) Decide whether TIRS application to waste processing is sufficiently viable to proceed with development.
5. Procure and modify needed instrumentation for developing TIRS for salt/polymer stream analysis. Performance Period: January - August 1993. Completed.
6. Develop TIRS for salt/polymer stream analysis on the laboratory scale. Performance Period: March - January 1994.
7. Develop and test rapid, minimum handling method for determining off-line the gross composition of the salt waste prior to encapsulation. Performance Period: April 1993 - April 1994.
8. Develop components for pilot scale tests at Brookhaven. Performance Period: September 1993 - January 1994.
FY94
9. Final TTP submittal. Performance Period: September 30, 1993. Completed.
10. Weekly highlights. Performance Period: October 10, 1993. Completed.
11. Monthly reports. Performance Period. November 10, 1993. Completed.
Subtask 1
12. Coordinate installation at Hanford of one of the FTIR-PAS systems from Activity No. 1. Transfer to Hanford the methodology from Activity No. 3 and the analysis data so far completed in Activity No. 2 through a training session by Ames Laboratory staff at Hanford. Performance Period: January 1994.
Milestone B. (Key Milestone) Determine that the methodology and sufficient FTIR-PAS expertise have been successfully transferred to Hanford staff so that analogous experimentation can begin at Hanford.
13. Continue transference of analysis expertise developed in Activity No. 2 to Hanford principally via correspondence and provision of data and algorithms. Performance Period: Feburary 1994 - September 1995.
14. Develop automatic sampling and shuttle system for use in hot environments. Performance Period: March 1994 - July 1995.
Subtask 2
15. Install TIRS on pilot-scale extruder and continue improving design. Performance Period: January - March 1994.
16. Demonstrate TIRS on pilot scale processing unit during encapsulation technology demonstration at Brookhaven National Laboratory. Performance Period: March 1994.
Milestone C. Select between heating and cooling versions of TIRS for continued development and scale up.
17. Continue development of chosen TIRS version, incorporating hot tests as required. Performance Period: April 1994 - July 1995.
18. Design and build low cost, filter-based TIRS monitor. Performance Period: May 1994 - September 1995.
FY95
Subtask 1
19. (Milestone D) Transfer sampling and shuttle system to Hanford. Priority: 100% previous year funding.
Subtask 2
20. (Milestone E) Deliver filter-based unit to user site and instruct site staff on its operation and maintenance. Priority: 100% previous year funding.
21. Prepare R&D Technical Status Report. Priority: 100% previous year funding.
Funding Basis
Operating Funds
The requested operating budgets are $300K for FY94 and $300K for FY95. The combined subtasks will require 3.5 full-time equivalents. Materials and supplies will cost $30K each year. This includes materials for developing the sampling shuttle and the simplified TIRS unit. $13K in FY94 and $14K in FY95 are budgeted for travel, which includes the installation and training sessions at Westinghouse Hanford and the TIRS installation and demonstration at Brookhaven National Laboratory.
Capital Funds
No capital funds are requested for FY94 and FY95.
Technology Need
Subtask 1
The wastes accumulated in storage tanks at the Hanford Site are one of the most urgent remediation problems in the DOE complex. The wastes are heterogeneous, highly concentrated and chemically active. A wide variety of analytical approaches will be required to produce a clear picture of the constituents and their distributions in the tanks. The sludges and solids are among the most difficult materials in the tanks to analyze. They are both more heterogeneous than the liquid phases, and their analysis by conventional means is more involved, since sludge and solid samples would normally require extensive preparation prior to their analysis. The FTIR-PAS technique that is proposed for application in this project is uniquely capable of spectroscopically analyzing solids and semisolids like the wastes with virtually no sample preparation and hence a minimum of handling. The amount of material necessary for FTIR-PAS analysis is quite small (typically one or a few milligrams), which also reduces the hazards and complexities of the analysis. FTIR-PAS will provide a quantitative compositional analysis of the multi-atomic ions and molecular species in the sludges and solids. These species are of special interest because they are the chemically active constituents in the wastes. FTIR-PAS is a general-purpose approach to the spectroscopic analysis of solids that requires little sample preparation and little sample volume. It is applicable to other sites or problems where solid or semi-solid samples must be analyzed while keeping sample handling or sample size to a minimum.
Subtask 2
Many of the present cementation processes being used to immobilize radioactive salts and similar wastes produce waste forms that fail to meet new RCRA requirements and are not certifiable. The polymer encapsulation process being developed at Brookhaven National Laboratory, Westinghouse Hanford and Rocky Flats is a response to this immediate need for a process that will produce waste forms that can be certified, transported, and disposed of properly. The monitoring system proposed in this task plan will both improve the quality of the waste form produced and provide documentation of its composition. The real-time information provided by the monitoring system will allow the waste-process operators to maintain the optimum salt-to-polymer ratio in the final waste form. This same information can act as documentation of the waste-form composition for certification purposes.
Alternatives
Subtask 1
Because of the chemical and radiological hazards presented by radioactive tank wastes, only those techniques which can analyze the wastes in their native state, without significant preparative handling or the addition of solvents, are viable alternatives. There are two spectroscopic techniques other than PAS, diffuse reflectance and attenuated total reflectance, that can be used on opaque pastes and therefore might be used on the solids and sludges. They, like PAS, have been developed in recent years for the spectroscopic analysis of materials that are opaque and cannot readily be made semi-transparent by physical thinning or dissolving. Diffuse reflectance is best suited for materials in which light can penetrate well into a sample and be scattered from many surfaces, as is the case with weakly absorbing or diluted powders, pastes and suspensions. Unfortunately, this makes diffuse reflectance poorly suited for the strongly absorbing and highly variable wastes. Attenuated total reflectance involves spreading the sample on a transparent crystal and making intimate contact between the sample and the crystal. Light is then reflected off the sample-crystal interface. This technique could work with sludges of the correct consistency, but the sample must generally be a gram or more in size and the required crystal is not robust, either physically or chemically. It will be attacked by the chemically active and abrasive materials in the sludges, destroying the calibration of the instrument.
Work has been done in developing near-infrared spectroscopy for the analysis of opaque, intractable materials. This region of the spectrum has the advantage that near-infrared light penetrates further into normally opaque materials, which reduces the problem of opacity, making it easier to acquire a spectrum from normally intractable samples. The disadvantage of the near-infrared is the severe overlapping of features. Even for simple samples, there is usually no one-to-one correspondence between spectrum features and chemical species as there is in the mid-infrared. As a result, near-infrared spectra are generally easier to acquire by conventional methods than mid-infrared spectra are, but the spectra are then harder to analyze. Near-infrared spectroscopy may at first glance be thought an alternative technique to FTIR-PAS, but in fact, the near-infrared is included within the FTIR-PAS approach. Although the initial thrust of the work will be aimed in the mid-infrared, the equipment to be procured can acquire spectra to out 8000 cm¯¹ (the middle of the near-infrared) and can be extended to include all of the near-infrared with relatively minor modifications. The near-infrared option will be kept open during the project.
Subtask 2
The present method used by those developing the encapsulation process to determine the salt-to-polymer ratio is off line. The density of the final waste form is measured, and the ratio of salt to polymer is calculated from the known densities of the pure salt and polymer. Although this is sufficient during the development process, it has the drawbacks of not being real time, of averaging over the complete waste form, and of assuming a constant salt density. Since the analysis is not real time, the results can not facilitate real time control of the processing. Since it measures the average of the complete waste form, it misses any fluctuations in the salt/polymer ratio that might signal poor encapsulation in part of the product. Lastly, it assumes the density of the salt is invariant. It is unlikely that the salt waste will be that well characterized during the actual production process. Its density may vary substantially, resulting in a corresponding variation in the calculated salt-to-polymer ratio.
The proposed monitoring system avoids all of these shortcomings. Its on-line, real-time response will allow the operators to adjust the processing to maintain an optimum ratio. It does not average over large volumes of material (unless desired), so it can spot fluctuations in the output. It need not assume any properties of the salt are constant. The polymer and different forms of salt (e.g., nitrates, sulfates, phosphates) are observed separately by the monitor, thus not only the density but the composition of the salt feedstock can vary without causing the monitor system to produce inaccurate results.
No other comparable technology is available for on-line, real-time analysis of the salt/polymer stream. Most commercially available on-line monitors measure a single physical property of the passing material stream and then calculate the corresponding composition, much as the off-line measurement of density is presently being used. This is not an acceptable approach here because of the variable nature of the salt-waste feedstock. The variable nature prevents there being a reliably quantitative relationship between the physical property and the salt-to-polymer ratio. The few spectroscopically based monitoring systems available force the material stream through a thin cell and pass light through the stream in order to observe a conventional transmission spectrum. This works with clear, low-viscosity streams of some polymers, but the milky, high-viscosity stream that results from mixing salt into the polymer is not suitable for such an approach. This is especially true at the high salt loadings planned.
Benefits
Subtask 1
The tank sludges and solids are proving themselves to be difficult materials to chemically analyze without substantial sample handling. The FTIR-PAS technology proposed for application in this project can rapidly provide spectra of the wastes with a minimum of handling. The FTIR-PAS spectra contain structure that is indicative of the kind and amount of chemically active species present in the wastes, thus the FTIR-PAS approach should rapidly provide badly needed data on the chemical environment in the tanks. The resulting benefits are many. The reduced sample handling of the approach will reduce the health hazards and difficulties of analysis. The increased efficiency and rapidity of analysis can improve remediation schedules. No solvents and only a small amount of sample (one to a few milligrams) are required, which minimizes the amount of waste requiring disposal after analysis. All of these benefits contribute to reduction in cost.
Subtask 2
The proposed monitoring system will improve operations, reduce costs, and minimize waste because it will act to improve the quality of the final waste form in a number of ways that the present off-line, density-based measurement cannot. The real-time measurement of composition by the monitoring system will allow the process operators to maintain the optimum composition in the product stream. This prevents the production of waste that is poorly encapsulated because of a too high salt-to-polymer ratio. Such waste would have to be reprocessed. It also prevents production of excessive volumes of the final waste form from a too low ratio. The off-line technique measures the average of the complete waste form and so misses any fluctuations in the salt/polymer ratio that might signal poor encapsulation in part of the product. The monitoring system will be able to track such fluctuations and alert the operators. The off-line measurement assumes the density of the salt is invariant. It is unlikely that the salt waste will be that well characterized during the actual production process. The monitoring system makes no such assumption. Lastly, the monitoring system will help in the waste certification process by providing documentation on the composition of the final waste form.
Criteria for Success
The success of this R&D project will be judged by the accuracy, reliability and user-acceptability of the measurements and methodologies. In Subtask 1, the accuracy of the methodology will be assessed by comparing the analysis results for a series of waste surrogates prescribed or provided by Westinghouse Hanford to their known compositions. The only uncertainty in this measure of success is how well the surrogates represent the actual wastes. The acceptability will be determined by whether personnel at Westinghouse Hanford wish to adopt the methodology. In Subtask 2, the accuracy and reliability of measurements will be judged by the performance of the TIRS system on the pilot-scale waste-processing system. The measurements must be accurate within the tolerances required by the waste-process development staff. They must also be reliable by being sufficiently stable to (or predictably dependent on) routine variations in the waste feedstock and in the waste-processing-system operating conditions. One of the ultimate goals is that the data from the measuring system be accepted as part of the documentation necessary for certifying waste forms, but fulfilling the extra requirements of this goal will not be addressed until the DT&E phase.
There are no supporting technologies in Subtask 1 that require development or are otherwise unavailable. In Subtask 2, the project is dependent on the development of the waste-processing lines at the user sites. The proposed schedule of milestones may need to be altered as the project proceeds to match changes at the user sites.
Regulatory Requirements
All of Subtask 1 and most of Subtask 2 only involve work with non-radioactive, surrogate waste materials, so no regulatory approvals, permits, or NEPA documentation requirements are anticipated. The hot work in the latter stages of Subtask 2 will be done at one of the user sites in conjunction with host personnel, so it will fall under the regulations in effect there at that time. Advancing the PAS approach in Subtask 1 to the regulator-accepted level will take place in outyears under a separate project. Should the technology in Subtask 2 advance to a DT&E phase, the necessary regulatory requirements will have to be met at that time.
Technology Transfer
Technology transfer as an intrinsic part of the project. In Subtask 1, installation trips and training sessions at Hanford are planned to explicitly transfer the technology. Frequent communication is planned between Ames and Hanford both to transfer the developed expertise from Ames and to receive feedback from Hanford on the details of the analysis problems and the acceptability of the solutions offered. The instrumentation at Westinghouse Hanford will be as identical to that used at Ames Laboratory as practicable so as to insure the transferability of the methodology, calibrations, and algorithms. Subtask 2 will be closely tied to the efforts at the user sites to develop the polymer encapsulation process so as to insure that the technology is tailored to user needs and is ready for implementation when required. Several on-site demonstrations and delivery of and training on the simplified TIRS unit are explicitly included in the milestones.
Acceptability
Acceptability to the user is one of the major measures of the success of the project. All activities in this project will be performed while in frequent communication with personnel at the user sites. At all points in Subtask 1, the methodology, algorithms, and data generated will be tailored to the needs identified by Hanford personnel. Activities in Subtask 2 will be integrated with the polymer-encapsulation-process development occurring at the user sites. This will insure that the technology is acceptable to the final users. Meeting the requirements of regulators will be the concern of separate projects in the outyears.
Environmental Safety and Health
All activities connected with this project are carried out in accordance with DOE environmental, safety, and health regulations. Those tasks done at Ames Laboratory are performed according to the Ames Laboratory Safety Manual, the Ames Laboratory Chemical Hygiene Plan, and the Environmental Technologies Development Program Unit Operations Manual. All research-group members have received the standard training for laboratory workers at Ames Laboratory as well as group-specific training. Three group members have taken the 40-hour HAZWOPER course (OSHA regulation 29 CFR 1910.120). Costs for personnel protective devices and other safety-related items used at Ames Laboratory are included in the figures presented in the Budget Expense Schedule by Cost Element attached to the end of this task plan. Tasks at demonstration sites are carried out in accordance with regulations and standard operating procedures in place at the sites.
Additional Items Required by the Characterization, Monitoring, and Sensor
Technologies Integrated Program
References
[1.] J. F. McClelland, R. W. Jones, S. Luo, and L. M. Seaverson, ``A Practical Guide to FTIR Photoacoustic Spectroscopy,'' in Practical Sampling Techniques for Infrared Analysis; P. B. Coleman, Ed.; CRC Press: Boca Raton, FL, 1993.
[2.] P. D. Kalb and P. Colombo, ``Polyethylene Solidification of Low-Level Wastes,'' Brookhaven National Laboratory Topical Report No. BNL 51867, U. S. Department of Energy, October 1984.
[3.] P. D. Kalb, J. H. Heiser III, and P. Colombo, ``Polyethylene Encapsulation of Nitrate Salt Wastes: Waste Form Stability, Process Scale-Up, and Economics ,'' Brookhaven National Laboratory Technology Status Topical Report No. BNL 52293, U. S. Department of Energy, July 1991.
[4.] R. W. Jones and J. F. McClelland, ``Quantitative Analysis of Solids in Motion by Transient Infrared Emission Spectroscopy Using Hot-Gas Jet Excitation,'' Analytical Chemistry, 1990, 62, 2074-2079.
[5.] R. W. Jones and J. F. McClelland, ``Transient Infrared Transmission Spectroscopy,'' Analytical Chemistry, 1990, 62, 2247-2251.
[6.] R. W. Jones and J. F. McClelland, ``Transient IR Spectroscopy: On-Line Analysis of Solid Materials,'' Spectroscopy, 1992, 7(4), 54-58.
[7.] J. F. McClelland and R. W. Jones, ``TIRS Shows Promise for On-Line QC,'' PI Quality, 1991, 1(2), 23-24.
[8.] R. W. Jones and J. F. McClelland, ``On-Line Analysis of Solids and Viscous Liquids by Transient Infrared Spectroscopy,'' Process Control and Quality, 1993, 4, 253-260.
Resumes
John McClelland.
Education: Ph.D. in Physics, 1976, Iowa State University; B.S. in Physics, 1965,
Dickinson College. Experience: Physicist, 1987-present, Ames Laboratory; Associate
Physicist, 1981-1986, Ames Laboratory; Postdoctoral Fellow, 1977-1980, Ames
Laboratory; Technical Staff, 1976-1977, Honeywell Electro-Optical Center;
Physicist, 1964-1970, National Bureau of Standards. Awards: Fellow of the
Optical Society of America; IR 100 Award, 1985; R&D 100 Award, 1992.
Roger Jones. Education: Ph.D. in Physical Chemistry, 1980, Massachusetts Institute of Technology; B.S. in Chemistry, 1975, University of California at Berkeley. Experience: Associate Chemist, 1988-present, Ames Laboratory and the Center for Advanced Technology Development, Iowa State University; Assistant Professor of Chemistry, 1982-1987, University of Alabama; Postdoctoral Researcher, 1980-1982, Cornell University. Awards: R&D 100 Award, 1992.
Cost/Benefit Analysis
The approach being taken to the analysis of the tank wastes at Hanford has been in flux. Accordingly, there is no single analysis approach that can act as a baseline against which the FTIR-PAS technology in Subtask 1 can be compared. The FTIR-PAS approach is faster, simpler, and requires less tank-waste handling than all other alternative approaches, thus the project technology should be favored in any cost/benefit analysis. The goals in Subtask 2 do not lend themselves to cost/benefit analysis. Since the two purposes of the TIRS monitor are the avoidance of mistakes during the operation of technology that is still in development and the provision of required documentation for waste-form certification, there are no explicit monetary costs against which to weigh the benefits of the TIRS technology.
Request-for-Proposals Needs
Subtask 1 addresses Need 4.3 (Characterization of underground storage tank waste) of the CMST-IP Request for Proposals, and Subtask 2 addresses Need 5.2 (On-line monitoring of waste treatment and remediation processes).
Specifications and Capabilities of Best Current Technology
There is no alternative analysis technique that requires both so small a sample volume to analyze and so little sample preparation as the FTIR-PAS technology used in Subtask 1. Under the proper conditions, ion chromatography can be more precise than FTIR-PAS, but it has several drawbacks. In ion chromatography the radioactive sample is dissolved in a solvent and certain of its properties (e.g., ionic strength, pH) must be adjusted before the chromatographic separation is started. The separation then normally typically takes roughly an hour. By contrast, FTIR-PAS only requires that a small amount of sample in its native (unprocessed) form be placed in the instrument. The analysis then requires 5 to 20 minutes, depending on difficulty. Accordingly, ion chromatography is substantially slower, requires much more sample handling, and generates far more hazardous waste. Two alternative spectroscopic techniques, diffuse reflectance and attenuated total reflectance (ATR), can produce results equivalent to PAS for the proper samples, but they are not applicable to tank sludges and solids. Diffuse reflectance works well with powdered, weakly absorbing materials, thus the tank wastes would have to be dried and diluted in a transparent powder matrix before they could be analyzed. ATR requires that the sample be smeared onto the surface of a special crystal, which is not physically or chemically robust, so ATR requires a much larger sample than FTIR-PAS and the tank wastes would attack the ATR crystal.
There is no alternative on-line monitor that could substitute for the TIRS technology of Subtask 2. Off-line analysis techniques can be made more precise than TIRS, but such techniques would be far too slow to contribute to the maintenance of process safety and control. On-line monitoring is required whenever the production of out-of-specification material can not be tolerated. Most commercially available on-line monitoring techniques determine a single physical property (e.g., color, temperature, reflectivity) of the stream and determine the desired property (salt-waste loading, in this case) from this by a predetermined relation between the two. This is acceptable on manufacturing lines where the feedstock and product streams are not expected to vary over time, but it is not practical here because of the variable nature of the salt waste. Such an approach would require that a complete calibration be done each time a new batch of waste is processed. Commercial infrared-spectroscopic analyzers do exist, but these are all based on transmission spectroscopy, in which the flowing process stream is forced between two closely spaced windows and light is passed through the thinned stream so that the amount of light transmitted through the stream may be measured. This works with clear, non-viscous streams, including many molten polymer process streams, but it is not applicable to the milky, highly viscous stream that results when salt and polymer are mixed.
Customer Data Quality Objectives
Customer Data Quality Objectives have not yet been defined. The Hanford tank sludges and solids dealt with in Subtask 1 are so variable and ill defined that no specifications for a minimally acceptable analysis have been drawn up. In Subtask 2 the polymer encapsulation process is still being developed, so no precise set of analysis specifications can be prepared. It is our intent to meet whatever requirements the users have, once they are able to define them.
Target Specifications and Capabilities
The overall target capability for the FTIR-PAS technology in Subtask 1 is the ability to quantify all important, non-trace, multi-atomic species in the all types of Hanford tank sludges and solids. No specific precision target has been established, since Data Quality Objectives have not been defined by the customer. The general target for the TIRS on-line monitor in Subtask 2 is the ability to determine in real-time (i.e., in under 2 minutes) the waste loading in the salt/polymer stream without contacting the stream or otherwise interfering with the processing. More precise targets will be established as the waste processing technology itself develops.
Integrated Demonstrations and Programs to Which this TTP has been Submitted
As this is a continuing project, rather than a new one, this TTP is being submitted only to the CMST-IP.