Search

This paper describes the experiences of the DIII-D programming staff in adapting Linux based Intel computing hardware for use in real-time data acquisition and feedback control systems. Due to the highly dynamic and unstable nature of magnetically confined plasmas in tokamak fusion experiments, real-time data acquisition and feedback control systems are in routine use with all major tokamaks. At DIII-D, plasmas are created and sustained using a real-time application known as the digital plasma control system (PCS). During each experiment, the PCS periodically samples data from hundreds of diagnostic signals and provides these data to control algorithms implemented in software. These algorithms compute the necessary commands to send to various actuators that affect plasma performance. The PCS consists of a group of rack mounted Intel Xeon computer systems running an in-house customized version of the Linux operating system tailored specifically to meet the real-time performance needs of the plasma experiments. This paper provides a more detailed description of the real-time computing hardware and custom developed software, including recent work to utilize dual Intel Xeon equipped computers within the PCS.

Purpose This phase I study was conducted to determine the toxicities, pharmacokinetics, and recommended doses of pemetrexed in cancer patients with normal and impaired renal function. Patients and Methods Patients received a 10-minute infusion of 150 to 600 mg/m2 of pemetrexed every 3 weeks. Patients were stratified for independent dose escalation by measured glomerular filtration rate (GFR) into four cohorts ranging from ≥ 80 to less than 20 mL/min. Pemetrexed plasma and urine pharmacokinetics were evaluated for the first cycle. Patients enrolled after December 1999 were supplemented with oral folic acid and intramuscular vitamin B12. Results Forty-seven patients were treated with 167 cycles of pemetrexed. Hematologic dose-limiting toxicities occurred in vitamin-supplemented patients (two; 15%) and nonsupplemented patients (six; 18%), and included febrile neutropenia (four patients) and grade 4 thrombocytopenia (two patients). Nonhematologic toxicities included fatigue, diarrhea, and nausea, and did not correlate with renal function. Accrual was discontinued in patients with GFR less than 30 mL/min after one patient with a GFR of 19 mL/min died as a result of treatment-related toxicities. Pemetrexed plasma clearance positively correlated with GFR (r2 = 0.736), resulting in increased drug exposures in patients with impaired renal function. With vitamin supplementation, pemetrexed 600 mg/m2 was tolerated by patients with a GFR ≥ 80 mL/min, whereas patients with a GFR of 40 to 79 mL/min tolerated a dose of 500 mg/m2. Conclusion Pemetrexed was well tolerated at doses of 500 mg/m2 with vitamin supplementation in patients with GFR ≥ 40 mL/min. Additional studies are needed to define appropriate dosing for renally impaired patients receiving higher dose pemetrexed with vitamin supplementation.

During the investigation of aviation accidents, postmortem samples from victims are submitted to the Federal Aviation Administration's Civil Aerospace Medical Institute for toxicological analysis. This report presents a rapid and reliable method for the identification and quantitation of sildenafil (Viagra) and its active metabolite, UK-103,320. This procedure utilizes sildenafil-d(8) as an internal standard for more accurate and reliable quantitation. The quantitation of sildenafil and UK-103,320 in postmortem fluid and tissue specimens obtained from victims from six separate aviation fatalities is described. The method incorporates solid-phase extraction and liquid chromatography-tandem mass spectrometry (LC-MS-MS) and MS-MS-MS utilizing an atmospheric pressure chemical ionization ion trap MS in the positive chemical ionization mode. Solid-phase extraction provided an efficient sample extraction yielding recoveries ranging from 79 to 88%. The limits of detection for sildenafil and UK-103,320 were 0.39 and 0.19 ng/mL, respectively. The linear dynamic range for both compounds was 0.78-800 ng/mL. The current method proved to be simple, accurate, and robust for the identification and quantitation of sildenafil and UK-103,320 in postmortem fluids and tissues.

Plasma shape control using real-time equilibrium reconstruction has been implemented on the National Spherical Torus Experiment (NSTX). The rtEFIT code originally developed for use on DIII-D was adapted for use on NSTX. The real-time equilibria provide calculations of the flux at points on the plasma boundary, which are used as input to a shape control algorithm known as isoflux control. The flux at the desired boundary location is compared with a reference flux value, and this flux error is used as the basic feedback quantity for the poloidal field coils on NSTX. The hardware that comprises the control system is described, as well as the software infrastructure. Examples of precise boundary control are also presented.

A complete software suite for integrated tokamak plasma control has been developed within the DIII-D program. The suite consists of software for real-time control of all aspects of the plasma, modeling, simulation and design tools for analysis and development of controllers, a flexible and modular architecture for implementation and testing of algorithms and many fully validated models. Many elements of the system have been applied to and implemented on NSTX and MAST. The DIII-D realtime plasma control system together with the integrated modeling and simulation suite have been selected for operational use by both the KSTAR and EAST tokamaks, and are also being used at General Atomics to investigate control issues for ITER.

The DIII-D mission to explore the advanced tokamak (AT) regime places significant demands on the DIII-D plasma control system (PCS) [D.A. Humphreys, et al., Fusion Eng. Des. 66–68 (2003) 633], including simultaneous and accurate regulation of plasma shape, stored energy, and density, as well as coordinated suppression of magnetohydrodynamic (MHD) instabilities. The present work describes selected new solutions to key control problems in DIII-D AT operation. These include novel nonrigid, resistive linear plasma response models, nonlinear algorithms for avoidance of current limiting, and improved neoclassical tearing mode control algorithms.

This paper describes the improvements being made in the capabilities of the DIII–D plasma control system (PCS) towards achieving optimization of current and pressure profiles in advanced tokamak discharges. Key improvements have been increased processing power and the ability to include profile diagnostic data. The recently completed upgrade of the PCS to Linux based Intel computers connected with 2 Gigabit/s Myrinet networking technology has been successful in achieving the goals of increasing the overall performance and flexibility of the system. The new Intel computing system has increased processing power by a factor 30 over the older i860 based systems. The Myrinet fiber based network has allowed the inclusion of data in real-time from DIII–D diagnostics, situated in remote locations within the DIII–D research facility. The PCS now collects 32 channels of motional stark effect (MSE) data and uses these data for real-time computation of the safety factor (q) profile. Electron temperature and density profile data from the Thomson scattering diagnostic and electron temperature profile data from the electron cyclotron emission (ECE) diagnostic are presently being added. Addition of an ion temperature and toroidal rotation profile data from the charge exchange recombination diagnostic is planned. Feedback control by the PCS of the electron temperature at two points has been demonstrated using either ECH or neutral beam power. This has been used to modify current profile evolution during plasma current ramp up. Specifics of the latest improvements to the DIII–D PCS are detailed here.

The integrated plasma control approach provides a systematic method for designing plasma control algorithms with high reliability and for confirming their performance off-line prior to experimental...

The DIII-D tokamak is capable of supporting a wide variety of plasma equilibria because of its relatively large number of coils and their proximity to the plasma. To support its advanced tokamak mission, the DIII-D experimental program continues to push the envelope of this capability, frequently encountering limits imposed by allowable currents in poloidal shaping coils. Violation of current constraints is presently dealt with by operator adjustment of control targets and gains between plasma discharges. At the same time, demands for more precise and stable control have motivated efforts to develop and install advanced multivariable algorithms for control of plasma shape in DIII-D and other devices. There is currently no way to ensure respect of nonlinear current constraints in a multivariable linear controller design and no practical way to manually tune these fully coupled controllers between discharges after installation. Various linear minimization schemes can be implemented to encourage currents to remain within limits, but adherence to these limits cannot be guaranteed by linear methods alone. In this paper, we describe ongoing efforts to provide methods that guarantee currents will not exceed preset limits, and that simultaneously achieve the best obtainable quality of control subject to current limit constraints.

An extensive set of software tools for integrated plasma control, developed and validated on the DIII-D tokamak, has been applied to several next-generation fusion device designs including KSTAR, EAST, and ITER. These devices will require elements of integrated plasma control in order to achieve high reliability advanced tokamak or burning plasma operation. Plasma Control Systems (PCS) based on the DIII-D PCS [1] have been designed for each of these devices. The integrated plasma control approach uses validated physics models to design controllers for plasma shape and both axisymmetric and nonaxisymmetric MHD instabilities [2] and confirms control performance by operating actual machine control hardware and software against detailed tokamak system simulations. The physics-based models include conductors, diagnostics, power supplies, and both linear and nonlinear plasma models. These models can be implemented in the detailed control simulations to verify event handling and demonstrate functioning of control action under realistic hardware (CPU and network) conditions. Results of simulations are shown, illustrating control performance characteristics produced by each device design, engineering choices, and control system algorithms and hardware. Such simulations allow confirmation of performance prior to actual implementation on an operating device.

Specimens from fatal aviation accident victims are submitted to the FAA Civil Aerospace Medical Institute for toxicological analysis. During toxicological evaluations, ethanol analysis is performed on all cases. Care must be taken when interpreting a positive ethanol result due to the potential for postmortem ethanol formation. Several indicators of postmortem ethanol formation exist; however, none are completely reliable. The consumption of ethanol has been shown to alter the concentration of two major serotonin metabolites, 5-hydroxytryptophol (5-HTOL) and 5-hydroxyindole-3-acetic acid (5-HIAA). While the 5-HTOL/5-HIAA ratio is normally very low, previous studies using living subjects have demonstrated that the urinary 5-HTOL/5-HIAA ratio is significantly elevated for 11-19 h after acute ethanol ingestion. Recently, our laboratory developed and validated an analytical method for the simultaneous determination of both 5-HTOL and 5-HIAA in forensic urine samples using a simple liquid/liquid extraction and LC/MS/MS and LC/MS/MS/MS. In this previous work a 15 pmol/nmol serotonin metabolite ratio cutoff was established in postmortem urine, below which it could be conclusively determined that no recent antemortem ethanol consumption had occurred. In the current study this newly validated analytical method was applied to five ethanol-positive aviation fatalities where the origin of the ethanol present could not previously be conclusively determined. In four of the five cases examined the detected ethanol was demonstrated to be present due to postmortem microbial formation, and not consumption, even though some indication of ethanol consumption may have been present.

During the investigation of aviation accidents, postmortem samples obtained from fatal accident victims are submitted to the FAA's Civil Aerospace Medical Institute (CAMI) for toxicological analysis. During toxicological evaluations, ethanol analysis is performed on all cases. Many species of bacteria, yeast, and fungi have the ability to produce ethanol and other volatile organic compounds in postmortem specimens. The potential for postmortem ethanol formation complicates the interpretation of ethanol-positive results from accident victims. Therefore, the prevention of ethanol formation at all steps following specimen collection is a priority. Sodium fluoride is the most commonly used preservative for postmortem specimens. Several studies have been published detailing the effectiveness of sodium fluoride for the prevention of ethanol formation in blood and urine specimens; however, our laboratory receives blood or urine in approximately 70% of cases. Thus, we frequently rely on tissue specimens for ethanol analysis. The postmortem tissue specimens received by our laboratory have generally been subjected to severe trauma and may have been exposed to numerous microbial species capable of ethanol production. With this in mind, we designed an experiment utilizing unadulterated tissue specimens obtained from aviation accident victims to determine the effectiveness of sodium fluoride at various storage temperatures for the prevention of microbial ethanol formation. We found that without preservative, specimens stored at 4 degrees C for 96 h showed an increase in ethanol concentration ranging from 22 to 75 mg/hg (average 42 +/- 15 mg/hg). At 25 degrees C, these same specimens showed an increase ranging from 19 to 84 mg/hg (average 45 +/- 22 mg/hg). With the addition of 1.00% sodium fluoride, there was no significant increase in ethanol concentration at either temperature.

Cocaine (COC) is one of the most widely abused illicit drugs in America. COC abuse transcends all social, racial, and economic boundaries. Following the introduction in the mid-1980s of a new form of cocaine called crack, cocaine use has been on the rise. Because of its intense high, crack smoking has become very popular. Despite its popularity, crack smoking is a particularly dangerous form of COC use. Additionally, COC and ethanol are frequently used together, resulting in the formation of a biologically active molecule that is nearly as psycoactive as COC but produces a longer lasting and toxic effect. Demonstrating the presence or absence of COC and COC-related molecules in postmortem fluids and/or tissues can have serious legal consequences and may help determine the cause of impairment and/or death. We have developed a simple method for the simultaneous determination of COC and the COC metabolites benzoylecgonine, norbenzoylecgonine, ecgonine methyl ester, ecgonine, and norcocaine, as well as anhydroecgonine methyl ester (a unique byproduct of COC smoking), cocaethylene (a molecule formed by the concurrent use of COC and ethanol) and their related metabolites, anhydroecgonine, norcocaethylene, and ecgonine ethyl ester. This method incorporates a Zymark RapidTrace automated solid-phase extraction system, gas chromatography/mass spectrometry, and PFP/PFPA derivatives. The lower limits of detection ranged from 0.78 - 12.5 ng/mL, and the linear dynamic range for most analytes was 0.78 3200 ng/mL. The extraction efficiencies were from 26 - 84%, with the exception of anhydroecgonine and ecgonine, which were from 1 - 4%. We applied this method to 5 aviation fatalities. This method has proven to be simple, robust, and accurate for the simultaneous determination of COC and 11 COC metabolites in postmortem fluids and tissues.

In this paper, we present an approach to automatically detect high impact coding errors in large Java applications which use frameworks. These high impact errors cause serious performance degradation and outages in real world production environments, are very time-consuming to detect, and potentially cost businesses thousands of dollars. Based on 3 years experience working with IBM customer production systems, we have identified over 400 high impact coding patterns, from which we have been able to distill a small set of pattern detection algorithms. These algorithms use deep static analysis, thus moving problem detection earlier in the development cycle from production to development. Additionally, we have developed an automatic false positive filtering mechanism based on domain specific knowledge to achieve a level of usability acceptable to IBM field engineers. Our approach also provides necessary contextual information around the sources of the problems to help in problem remediation. We outline how our approach to problem determination can be extended to multiple programming models and domains. We have implemented this problem determination approach in the SABER tool and have used it successfully to detect many serious code defects in several large commercial applications. This paper shows results from four such applications that had over 60 coding defects.

This paper presents the latest status of the DIII-D real-time digital plasma control system (PCS). The primary focus will be on the new computing and data acquisition hardware recently incorporated into the PCS and the added features provided by the new hardware. The upgrade of the PCS real-time computing and data acquisition hardware has been in progress for the past three and a half years. Since the initial proposal to migrate off of the old versa module europa bus standard (VME) i860 based system to more modern computing hardware, the PCS has seen a number of changes and improvements occurring in several planned phases. The final phase of the upgrade was successfully completed earlier this year. This phase included the complete removal of the i860 computers and Traq digitizer hardware and the incorporation of high performance Intel peripheral component interconnect (PCI) based computers fitted with 32 Channel PCI digitizers from D-TACQ corporation. Distinguishing features of the completed PCS upgrade include a custom real-time data acquisition and control solution based on the Linux Intel computing platform and a scaleable multi CPU parallel processing architecture connected by a two gigabit per second Myrinet network. Among the many improvements resulting from the upgrade are the higher computing performance (factor 20 over the previous system), connectivity to remote DIII-D diagnostics through Myrinet fiber optic links and plasma diagnostic data displayed in real-time on digital oscilloscopes.

Toxicological examination of fatal aviation accident victims routinely includes analysis of ethanol levels. However, distinguishing between antemortem ingestion and postmortem microbial formation complicates all positive ethanol results. Development of a single analytical approach to determine concentrations of 5-hydroxytryptophol (5-HTOL) and 5-hydroxyindole-3-acetic acid (5-HIAA), two well-known metabolites of serotonin, has provided a convenient, rapid and reliable solution to this problem. Antemortem ethanol leads to an elevation in the 5-HTOL/5-HIAA ratio for 11-19 h after acute ingestion. The liquid-liquid extracts of postmortem urine samples were subjected to liquid chromatography-mass spectrometry (LC-MS) for the simultaneous quantitation of these two analytes, yielding detection limits of 0.1 ng/ml for each. Examination of the 5-HTOL/5-HIAA ratio was undertaken for 44 urine samples known to be antemortem ethanol-positive or antemortem ethanol-negative. Recent ethanol ingestion was conveniently and accurately separated using a 5-HTOL/5-HIAA ratio of 15 pmol/nmol, a value previously suggested using human volunteers. All 21 ethanol-negative postmortem samples were below this cutoff, while all 23 ethanol-positive postmortem samples were above this cutoff. Thus, we recommend the employment of this cutoff value, established using this straightforward LC-MS procedure, to confirm or deny recent antemortem ethanol ingestion in postmortem urine samples.

This paper describes a comprehensive set of technologies that enables rich interaction paradigms for Web applications. These technologies improve the richness of user interfaces and the responsiveness of user interactions. Furthermore, they support disconnected or weakly connected modes of operation. The technologies can be used in conjunction with many Web browsers and client platforms, interacting with a J2EE™ server. The approach is based on projecting the server-side model-view-controller paradigm onto the client. This approach is firmly rooted in the Web paradigm and proposes a series of incremental extensions. Most of the described technologies have been adopted by IBM server (WebSphere®) and client products.

During the investigation of aviation accidents, postmortem samples from accident victims are submitted to the FAA's Civil Aerospace Medical Institute for toxicological analysis. In order to determine if an accident victim was exposed to an in-flight/postcrash fire or faulty heating/exhaust system, the analysis of carbon monoxide (CO) is conducted. Although our laboratory predominantly uses a spectrophotometric method for the determination of carboxyhemoglobin (COHb), we consider it essential to confirm with a second technique based on a different analytical principle. Our laboratory encountered difficulties with many of our postmortem samples while employing a commonly used GC method. We believed these problems were due to elevated methemoglobin (MetHb) concentration in our specimens. MetHb does not bind CO; therefore, elevated MetHb levels will result in a loss of CO-binding capacity. Because most commonly employed GC methods determine %COHb from a ratio of unsaturated blood to CO-saturated blood, a loss of CO-binding capacity will result in an erroneously high %COHb value. Our laboratory has developed a new GC method for the determination of %COHb that incorporates sodium dithionite, which will reduce any MetHb present to Hb. Using blood controls ranging from 1% to 67% COHb, we found no statistically significant differences between %COHb results from our new GC method and our spectrophotometric method. To validate the new GC method, postmortem samples were analyzed with our existing spectrophotometric method, a GC method commonly used without reducing agent, and our new GC method with the addition of sodium dithionite. As expected, we saw errors up to and exceeding 50% when comparing the unreduced GC results with our spectrophotometric method. With our new GC procedure, the error was virtually eliminated.

The establishment of the Web application as a successful application model may be attributed to its adherence to open standards, to the powerful J2EE™ application framework for the execution of business and process logic, and to its integration with heterogeneous external systems, transactional processes, and Web technologies. In addition, effective tool support has enabled rapid development and deployment, powerful processes composed from components, and access to a steadily increasing range of function. Many Web applications are designed for high performance, scalability, and highly reliable utilization and are capable of extending modes of usage and growing throughput. Successful application development in these environments involves bridging the gap between exercising available programming specifications and the proper design, coding, and life-cycle management of the application. This paper addresses issues, transactional and nontransactional, in application development that are important for the successful deployment and execution of enterprise Web applications in production environments. These issues are based on experience with deployments of Web applications in various customer environments and are not new, but are either fundamental to previous transactional systems or have appeared in other classes of applications. Understanding these issues is essential both to building effective tooling to support the runtimes and to the evolution and definition of the runtimes themselves.

A271 ADVANCED TOKAMAK OPERATION USING THE DIII-D PLASMA CONTROL SYSTEM. The principal focus of experimental operations in the DIII-D tokamak is the advanced tokamak (AT) regime to achieve, which requires highly integrated and flexible plasma control. In a high performance advanced tokamak, accurate regulation of the plasma boundary, internal profiles, pumping, fueling, and heating must be well coordinated with MHD control action to stabilize such instabilities as tearing modes and resistive wall modes. Sophisticated monitors of the operational regime must provide detection of off-normal conditions and trigger appropriate safety responses with acceptable levels of reliability. Many of these capabilities are presently implemented in the DIII-D plasma control system (PCS), and are now in frequent or routine operational use. The present work describes recent development, implementation, and operational experience with AT regime control elements for equilibrium control, MHD suppression, and off-normal event detection and response.

A comprehensive investigation of electric field effects associated with the backside Ge profile in SiGe heterojunction bipolar transistors (HBTs) is conducted using calibrated simulations. We show for the first time that the backside Ge retrograde can alter the local electric field distribution in the base–collector space-charge region near the SiGe to Si heterojunction, thereby affecting the impact ionization and the apparent neutral base recombination (NBR) ( I B ( V CB )/ I B ( V CB =0)) of SiGe HBTs. The changes in the electric field induced by the Ge-induced band offsets contributes to a decrease of the observed impact ionization between comparably doped SiGe HBTs and Si bipolar junction transistors (BJTs), as well as an improved V CB dependence of I B (apparent decrease in NBR). Experimental data on SiGe HBTs with various Ge profiles and a Si BJT control are used to support our claims.

During the investigation of aviation accidents, postmortem specimens from accident victims are submitted to the Federal Aviation Administration's Civil Aerospace Medical Institute (CAMI) for toxicological analysis. A case recently received by CAMI screened positive for the anticonvulsant medication carbamazepine (Tegretol ) by gas chromatography/mass spectrometry (GC/MS). The carbamazepine found during the routine screening procedure was subsequently confirmed using a carbamazepine-specific GC/MS procedure. Concurrently, it was discovered that the accident victim had been prescribed oxcarbazepine (Trileptal ). Oxcarbazepine is nearly structurally identical to carbamazepine and is metabolized by cytosolic enzymes in the liver to the active compound 10,11-dihydro-10-hydroxycarbamazepine. The carbamazepine initially found in this case was present due to the breakdown of the active oxcarbazepine metabolite in the GC/MS injector port. In the current study, this conversion is investigated, the percentage of carbamazepine formed at various injector port temperatures is determined, and these three compounds are quantified in nine fluid and tissue specimens from the case in question. Finally, liquid chromatography/mass spectrometry (LC/MS) was used to demonstrate the absence of carbamazepine, and its formation, in the same specimens.

This paper describes an upgrade for the real-time computing system responsible for monitoring and controlling plasma, discharges in the DIII–D tokamak (J.L. Luxon, L.G. Davis, Fusion Technol. 8 (1985) 441) at General Atomics (GA). The current system employs six CSPI i860 VME format processors working in parallel to acquire data in real-time and perform feed back control of plasma shape and position parameters. Work has commenced on integration of a new computing system based on commonly available PCI bus based processors that communicate over the 2 Gbit/s Myrinet (Myricom, Inc., CA, USA) network. The new system will greatly improve the processing power available to the algorithms required for computing plasma equilibrium reconstructions in real-time. A factor of 20 anticipated performance increase will allow for improved accuracy and frequency response for plasma shape estimation and control. The migration from VME to PCI Myrinet computer clustering will improve the data acquisition capabilities by opening up access to additional DIII–D temperature and density diagnostics. The upgrade will increase the inter-processor communication speed and provide the flexibility to integrate additional processors to match the cost and computing needs of the tokamak research program.

This double-masked, randomized, placebocontrolled study was conducted in healthy adult male and female volunteers with no clinically relevant baseline electrocardiographic (ECG) abnormalities to assess the cardiac tolerability margin of sparfloxacin (as measured by the effect on QT c interval) under conditions of potential overdose at up to 4 times the usual therapeutic loading dose. The 23 enrolled volunteers received a sequence of single doses of sparfloxacin (400, 800, 1200, and 1600 mg), 1 dose in each of 4 study periods. Six volunteers received placebo during each period. A 14-day washout separated the periods. Serial blood samples and ECG measurements were collected in each period to determine the pharmacokinetic and pharmacodynamic characteristics of sparfloxacin. The area under the concentration-time curve from time zero to infinity (AUC 0–∞ ) exhibited dose proportionality. The maximum plasma concentration (C max ) after the 1200-and 1600-mg doses was lower than would be expected for a linear dose relationship. This was also the case with the mean increase and mean maximum increase in QT c interval. Increases in the QT c interval correlated well with C max but not with AUC 0–∞ . The time to reach C max showed a slight tendency to increase with dose, as did the terminal elimination half-life. Changes in QT c -interval dispersion were similar for both placebo recipients and sparfloxacin-treated volunteers and were of no clinical consequence. At supratherapeutic doses, the extent of sparfloxacin's absorption (AUC 0–∞ ) was dose independent; however, the rate of absorption was dose dependent, with C max increasing substantially less than proportionally to the administered dose. This limited the C max of sparfloxacin at supratherapeutic doses and thus the increase in QTC c interval. Rechallenge demonstrated that only 2 of 8 subjects had the same degree of QT c -interval prolongation, emphasizing the marked variability in the QT c interval.

Sparfloxacin, a broad-spectrum, oral fluoroquinolone antimicrobial agent, has a long elimination half-life that permits once-daily administration. Antibiotics may increase the oral bioavailability of digoxin, leading to increases in its plasma concentration. Since patients treated with sparfloxacin may be receiving concurrent treatment with digoxin, the possibility of an interaction between sparfloxacin and digoxin was examined in a doublemasked, placebo-controlled, multipledose, two-way crossover study in 24 healthy male volunteers between 20 and 49 years of age. All subjects were given digoxin 0.3 mg once daily throughout the 20-day study. Sparfloxacin (or placebo) was given as a 400-mg loading dose on day 1, followed by single 200-mg daily doses for 9 days, with crossover to the alternate treatment on days 11 through 20. Plasma levels of digoxin were analyzed by validated radioimmunoassay, and plasma levels of sparfloxacin were analyzed by validated high-performance liquid chromatography. Concomitant administration of sparfloxacin and digoxin was generally well tolerated. Mean values for steady-state area under the concentration-time curve over 24 hours for the 2 treatments were virtually identical: 28.4 ng/h per mL −1 for digoxin administered with placebo and 28.9 ng/h per mL −1 for digoxin administered concomitantly with sparfloxacin. Mean steady-state maximum plasma concentrations were 3.91 and 3.59 ng/mL for digoxin with placebo and digoxin with sparfloxacin, respectively. Mean steady-state trough plasma digoxin concentrations for the 2 treatments were 0.87 and 0.89 ng/mL, respectively. Mean times to steady-state maximum plasma concentrations were identical at 0.89 hours for both treatments. Mean steady-state oral clearance was 10.6 L/h for digoxin alone and 10.4 L/h for digoxin with sparfloxacin. Thus administration of sparfloxacin in combination with digoxin did not alter the pharmacokinetics of digoxin in healthy male volunteers aged 20 to 49 years. Steady-state plasma sparfloxacin concentrations were consistent with those obtained in other multiple-dose phase I studies, suggesting that digoxin does not alter the steady-state pharmacokinetics of sparfloxacin.

Sparfloxacin is a broad-spectrum oral fluoroquinolone antimicrobial agent with a long elimination half-life (t(1/2)). Concurrent treatment with antacids has demonstrated a reduction in the oral absorption of many quinolones. This study was undertaken to determine an optimal time for dosing antacids in relation to sparfloxacin administration to minimize antacid-induced reduction in sparfloxacin bioavailability. This open-label, single-dose, randomized, four-way crossover study was conducted in 20 healthy male volunteers between the ages of 18 and 38 years. Treatments consisted of single 400-mg oral doses of sparfloxacin alone and with Maalox 30 mL given 2 hours before, 2 hours after, and 4 hours after oral administration of sparfloxacin. All 20 subjects completed the study. A 400-mg single oral dose of sparfloxacin was well tolerated both given alone and with Maalox. Maalox given 4 hours after sparfloxacin administration was the only regimen that did not cause a statistically significant reduction in the rate and extent of sparfloxacin absorption. The 90% confidence intervals comparing sparfloxacin alone with the preceding regimen in terms of area under the concentration-time curve from zero to infinity (AUC(0-infinity)) and maximum plasma concentration (Cmax) were within the range 80% to 125%. Administration of Maalox 2 hours before, 2 hours after, and 4 hours after sparfloxacin caused mean decreases in AUC(0-infinity) of 23%, 17%, and 5%, respectively. Corresponding decreases in mean Cmax values were 29% with Maalox administered 2 hours before sparfloxacin and 13% with Maalox administered 2 hours after sparfloxacin. The mean Cmax value was un affected when Maalox was administered 4 hours after sparfloxacin administration. The 90% confidence intervals for these comparisons with sparfloxacin alone were outside the 80% to 125% range and did not include 100%. Time to Cmax and t(1/2) were similar for all four regimens. The elimination rate of sparfloxacin was unaffected by concomitant administration with Maalox in healthy male volunteers.

SUMMARY. The aims of this investigation were firstly to determine the level of agreement between angular measures of lumbar spinal motion recorded by the OSI CA 6000 Spine Motion Analyzer (OSI SMA) and measures obtained from X-rays of subjects without lumbar symptoms, and secondly to determine the intra-rater reliability of measures of lumbar spinal range of motion using the OSI SMA on individuals with lumbar symptoms. To fulfill the first aim 13 asymptomatic subjects (mean age = 28.6 years; SD = 7.1) performed the motions of lumbar flexion, extension, side bending to the right, and side bending to the left twice while wearing the OSI SMA, and once for the exposure of the X-ray. The difference between the measures obtained by the two methods was plotted against the average of those measures for each subject to illustrate the level of agreement of the two methods. Differences between measures that were within one standard deviation of the mean were as follows: for flexion, 7 out of 10 showed less than 8 degrees difference; for extension, 7 out of 10 showed less than 7 degrees difference; for side bending to the right, 11 out of 13 showed less than 6 degrees difference; and for side bending to the left, 10 out of 13 showed less than 5 degrees difference. For asymptomatic subjects, measures obtained from the OSI SMA appear to provide good agreement with measures of lumbar spinal motion in the sagittal and frontal planes as determined by X-rays. To fulfil the second aim of the investigation thirty subjects with lumbar symptoms (mean age = 35.9 years; SD = 14.2) performed four trials of lumbar motion to their maximum for flexion, extension, side bending to the right and left, and rotation to the right and left. The OSI SMA hardware was then completely removed and replaced by the same examiner, and the trials were repeated. ICC values were 0.903 or higher for all motions. These results suggest that active range of motion measures obtained with the OSI SMA in subjects with lumbar symptoms are consistent over repeated trials. Copyright 1997 Harcourt Publishers Ltd.

Electron-diffraction patterns of the fullerene C70 in the gaseous state at 810−835 °C have been recorded in experiments similar to those recently described for C60. The radial distribution curve calculated from the scattered intensity is entirely consistent with a molecule of D5h symmetry. With assumption of this symmetry, 12 parameters are required to specify the structure. Reliable values are thus much more difficult to obtain for these parameters than for C60 whose structure is completely defined by only two parameters. Six different models were found that give excellent fits to the diffraction data. The models may be divided into two types characterized either by a shorter (1.4+ A) or a longer (1.5+ A) equatorial bond. Despite this difference, however, the average length of the eight bonds is similar in all models (1.434 A; average deviation 0.006 A). Since no model could be favored on the basis of the electron-diffraction data alone, a best model was selected from considerations of theoretical energi...

During the tokamak discharge, especially the ramp-up phase, the plasma state equilibrium continually evolves. As a consequence, the plasma response model should evolve as well. We first identified a linear plasma response model of the rotational transform ι profile and βN around a desired equilibrium. Then, an uncertainty is introduced to the identified model to partially account for the dynamic character of the plasma state equilibrium evolution. A robust controller is designed to stabilize this family of plasma models, which are reformulated into a nominal model with uncertainty. A singular value decomposition (SVD) of the nominal identified model is carried out to decouple and identify the most relevant control channels in steady-state. The DK-iteration method, combining H∞ synthesis and μ analysis, is applied to synthesize a closed-loop controller that minimizes the tracking error and input effort. The feedback controller is then augmented with an anti-windup compensator, which keeps the given profile controller well-behaved in the presence of magnitude constraints in the actuators and leaves the nominal closed-loop unmodified when no saturation is present. PTRANSP simulations and experimental results in DIII-D illustrate the performance of the model-based controller.

The paper proposes a model-based control approach for the coupled evolution of the poloidal magnetic flux profile and the normalized pressure ratio βN. The model is determined by a system identification method which is shown to sufficiently reproduce the plasma response to variations in particular actuators. Data for model identification is collected during the plasma current flattop in a large βN, high-confinement scenario (H-mode) with the actuators modulated in open loop. Using this data, a linear state-space plasma response model for the poloidal magnetic flux profile and βN dynamics around a plasma equilibrium state is identified. An optimal state feedback controller with integral action is designed for the purpose of simultaneous control of the poloidal flux profile and βN. Experimental results showing the performance of the proposed controller implemented in the DIII-D tokamak are presented.

A number of important software and hardware changes have recently been made to the DIII-D Plasma Control System (PCS) to further its capabilities in support of fusion research. The PCS is a highly customizable real-time control application developed at General Atomics used to manage the many parameters that affect plasmas produced on the DIII-D tokamak. Included in the most recent updates to the PCS are refinements to the real-time Electron Cyclotron Heating (ECH) capabilities which have improved overall performance and reliability for fast and precise aiming of the mirrors used to control direct ECH power into the plasma. The introduction of new real-time streaming data acquisition hardware has provided a means for acquiring plasma electron temperatures and densities from the Thomson scattering System along with data from the Electron Cyclotron Emission (ECE) diagnostic for use in PCS feedback control algorithms. The new fiber optically connected streaming digitizers allow PCS computers located in one part of the tokamak facility to easily communicate with remotely located diagnostic systems in other parts of the lab, in addition to being able to transfer high frequency data (sampled at 500 Hz) for a large number of channels in real-time. Details of the most recent PCS enhancements will be provided along with a more thorough description of the latest software and hardware architecture.

The DIII-D Plasma Control System (PCS [1]), initially deployed in the early 1990's, now controls nearly all aspects of the tokamak and plasma environment. Versions of this PCS, supported by General Atomics, are presently used to control several tokamaks around the world, including the superconducting tokamaks EAST and KSTAR. The experimental challenges posed by the advanced tokamak mission of DIII-D and the variety of devices supported by the PCS have driven the development of a rich array of control algorithms, along with a powerful set of tools for algorithm design and testing. Broadly speaking, the PCS mission is to utilize all available sensors, measurements and actuators to safely produce a plasma state trajectory leading to and then maintaining the desired experimental conditions. Often new physics understanding leads to new or modified control requirements that use existing actuators in new ways. We describe several important DIII-D PCS design and test tools that support implementation and optimization of algorithms. We describe selected algorithms and the ways they fit within the PCS architecture, which in turn allows great flexibility in designing, constructing and using the algorithms to reliably produce a desired complex experimental environment. Control algorithms, PCS interfaces, and design and testing tools are described from the perspective of the physics operator (PO), who must operate the PCS to achieve experimental goals and maximize physics productivity of the tokamak. For example, from a PO's (and experimental team leader's) standpoint, a PCS algorithm interface that offers maximum actuator, algorithmic and measurement configuration flexibility is most likely to produce a successful experimental outcome. However, proper constraints that limit flexibility in use of the PCS can also help to maximize effectiveness. For example, device limits and safety must be built into the PCS, sometimes at the algorithm level. We show how the D3D PCS toolset enables rapid offline testing of a new or modified algorithm in a simulated tokamak environment. Finally, we illustrate usage of PCS-based checklists and procedures that enhance experimental productivity and we describe an asynchronous condition detector system within the PCS that enhances device safety and enables complex experiment design.

Sertraline (Zoloft) is a selective serotonin reuptake inhibitor that is a commonly prescribed drug for the treatment of depression, obsessive-compulsive disorder, panic disorder, social anxiety disorder, premenstrual dysphoric disorder and post-traumatic stress disorder. Although the use of sertraline is relatively safe, certain side effects may negatively affect a pilot’s performance and become a factor in an aviation accident. The authors’ laboratory investigated the distribution of sertraline and its primary metabolite, desmethylsertraline, in various postmortem tissues and fluids obtained from 11 fatal aviation accident cases between 2001 and 2004. Eleven specimen types were analyzed for each case, including blood, urine, vitreous humor, liver, lung, kidney, spleen, muscle, brain, heart and bile. Human specimens were processed utilizing solid-phase extraction, followed by characterization and quantitation employing gas chromatography – mass spectrometry. Whole blood sertraline concentrations obtained from these 11 cases ranged from 0.005 to 0.392 mg/mL. The distribution coefficients of sertraline, expressed as specimen/ blood ratio, were as follows: urine, 0.47+ 0.39 (n 5 6); vitreous humor, 0.02+ + 0.01 (n 5 4); liver, 74+ 59 (n 5 11); lung, 67+ + 45 (n 5 11); kidney, 7.4+ + 5( n 5 11); spleen, 46+ + 45 (n 5 10); muscle, 2.1+ + 1.3 (n 5 8); brain, 22+ + 14 (n 5 10); heart, 9+ 7 (n 5 11); and bile, 36+ 26 (n 5 8). Postmortem distribution coefficients obtained for sertraline had coefficients of variation ranging from 47– 99%. This study suggests that sertraline likely undergoes significant postmortem redistribution.

Studies carried out using engineered proteins have demonstrated that protein adsorption to functional surfaces involves multiple interactions between specific groups on the protein and complementary binding sites distributed on the surface. A consequence of multipoint interactions is that protein binding affinity should depend strongly on the distribution of surface binding sites. In this investigation we present a thermodynamic framework for multipoint protein binding to a random arrangement of surface binding sites that also includes lateral interactions among adsorbed protein molecules. This framework results in reversible adsorption behavior analogous to that predicted by the Temkin model and chromatographic behavior analogous to that predicted by the “stoichiometric displacement” model (SDM). Using this framework we can now interpret the semiempirical parameters obtained using these models for protein binding in chromatographic systems in terms of thermodynamic parameters for protein−surface interact...

System identification techniques have been successfully used to obtain linear dynamic plasma response models around a particular equilibrium in different tokamaks. This paper identifies a two-time-scale dynamic model of the rotational transform ι profile and β N in response to the electric field due to induction as well as to heating and current drive (H&CD) systems based on experimental data from DIII-D. The control goal is to regulate the plasma ι profile and β N around a particular target value. A singular value decomposition (SVD) of the plasma model at steady state is carried out to decouple the system and identify the most relevant control channels. A mixed sensitivity H ∞ control design problem is solved to determine a stabilizing feedback controller that minimizes the reference tracking error and rejects external disturbances with minimal control energy. The feedback controller is augmented with an anti-windup compensator, which keeps the given controller well-behaved in the presence of magnitude constraints in the actuators and leaves the nominal closed-loop unmodified when no saturation is present. Experimental results illustrate the performance of the proposed controller, which is one of the first profile controllers integrating magnetic and kinetic variables ever implemented in DIII-D.

In tokamak fusion plasmas, controlling the spatial distribution profile of the toroidal current is key to achieving advanced scenarios characterized by confinement improvement and possible steady-state operation. The dynamics of the current profile are nonlinear and coupled with other plasma parameters, motivating the use of model-based control strategies. In this work, we use a control-oriented model of the current profile evolution in DIII-D to design a feedback controller for regulating the profile around a desired trajectory. Without feedback, the response of the current profile to disturbances, model uncertainty, and perturbed initial conditions can be undesirable. To improve tracking performance of the system, a nonlinear input transformation is combined with a linear-quadratic-integral (LQI) optimal controller designed to minimize a weighted combination of the tracking error and controller effort. The resulting control law utilizes the total plasma current, total external heating power, and line averaged plasma density as actuators. We present simulation and experimental results showing successful rejection of perturbed initial conditions and input disturbances.

Objective To audit routine measurement of alpha-fetoprotein (AFP) and acetylcholinesterase in amniotic fluid (AF) samples obtained at second-trimester amniocentesis. Methods We reviewed retrospectively 1737 consecutive AF specimens obtained for cytogenetic evaluation over a 4-year period and routinely assayed for AFP and acetylcholinesterase. In all instances, high-resolution ultrasonography was performed before amniocentesis. Details of pregnancy outcome of all cases with AF AFP levels greater than 2.0 multiples of the median and a positive or faint acetylcholinesterase band were obtained. Results There were 31 abnormal results (1.8%, 1 of 56). Of these, 25 cases had elevated AF AFP and/or positive acetylcholinesterase. Ultrasonography correctly identified all 18 fetuses with anomalies associated with abnormal levels of these biochemical markers, including open neural tube defects and/or anterior abdominal wall defects (17 cases) and fetal hydrops (one). In the remaining seven, no fetal anomalies were detected, and all neonates were structurally normal after birth. In addition, six pregnancies with faint acetylcholinesterase and normal AF AFP showed no fetal abnormalities at ultrasonographic examination and postdelivery. Conclusions High-resolution Ultrasonography was more accurate than AF biochemistry in the detection of congenital anomalies associated with elevated AFP levels and acetylcholinesterase in the AF. Routine measurement of these biochemical markers in AF samples obtained for cytogenetic analysis appears to have a very low yield and would therefore not be cost-effective in practices where high-resolution Ultrasonography is performed before amniocentesis.

The tokamak is a high order, distributed parameter, nonlinear system with a large number of instabilities. Therefore, accurate theoretical plasma models are difficult to develop. However, linear plasma response models around a particular equilibrium can be developed by using data-driven modeling techniques. This paper introduces a linear model of the rotational transform ι profile evolution based on experimental data from the DIII-D tokamak. The model represents the response of the ι profile to the electric field due to induction as well as to heating and current drive (H&CD) systems. The control goal is to use both induction and H&CD systems to regulate the plasma ι profile around a particular target profile. A singular value decomposition (SVD) of the plasma model at steady state is carried out to decouple the system and identify the most relevant control channels. A mixed sensitivity H ∞ control design problem is formulated to synthesize a stabilizing feedback controller without input constraint that minimizes the reference tracking error and rejects external disturbances with minimal control energy. The feedback controller is then augmented with an anti-windup compensator, which keeps the given profile controller well-behaved in the presence of magnitude constraints in the actuators and leaves the nominal closed-loop unmodified when no saturation is present. Finally, computer simulations and experimental results illustrate the performance of the model-based profile controller.

Setting up a suitable toroidal current profile in a fusion tokamak reactor is vital to the eventual realization of a commercial nuclear fusion power plant. Creating the desired current profile during the ramp-up and early flat-top phases of the plasma discharge and then actively maintaining this target profile for the remainder of the discharge is the goal at the DIII-D tokamak. The evolution of the toroidal current profile in toka-maks is related to the evolution of the poloidal magnetic flux profile, which is modeled by the magnetic diffusion equation. A simplified first-principles-driven, nonlinear, dynamic, control-oriented, partial differential equation model of the poloidal flux profile evolution is obtained by combining the magnetic diffusion equation with empirical correlations obtained from experimental data at DIII-D and is used to synthesize a robust H ∞ feedback controller to track a desired reference trajectory of the poloidal magnetic flux gradient profile. We employ a singular value decomposition of the static gain matrix of the plant model to identify the most relevant channels which we control with the feedback controller. A framework for real-time feedforward + feedback control was implemented in the DIII-D Plasma Control System and experimental results in the DIII-D tokamak are presented to illustrate the capabilities of the feedback controller. These experiments mark the first time ever a first-principles-driven model-based magnetic profile controller was successfully implemented and tested in a tokamak device.

Control of the spatial profile of plasma current in tokamak plasmas has been demonstrated to be a key condition for achieving advanced scenarios with improved confinement and possible steady-state operation. The dynamics of the current profile are nonlinear and coupled with several other plasma parameters, motivating the design of model-based controllers that can account for these complexities. In this work, we consider a control-oriented model of the current profile evolution in DIII-D and the problem of regulating the current profile around a desired feed-forward trajectory. In open-loop, the response of the system to disturbances and perturbed initial conditions may be undesirable. To improve the performance of the system, the PDE model is discretized in space using a finite difference method and a backstepping design is applied to obtain a discrete transformation from the original system into an asymptotically stable target system with desirable properties. Through a nonlinear transformation, the resulting boundary control law utilizes the total plasma current, total power, and line averaged density as actuators. A Simserver simulation study is done to test the controller's performance and its implementation in the DIII-D plasma control system. Finally, experimental results showing the ability of the controller to reject input disturbances and perturbations in initial conditions are presented.

First-principles predictive models based on flux-averaged transport equations often yield complex expressions not suitable for real-time control implementations. It is however always possible to reduce these models to forms suitable for control design while preserving the dominant physics of the system. If further model simplification is desired at the expense of less model accuracy and controller capability, data-driven modeling emerges as an alternative to first-principles modeling. System identification techniques have the potential of producing low-complexity, linear models that can capture the system dynamics around an equilibrium point. This paper focuses on the control of the poloidal magnetic flux profile evolution in response to the heating and current drive (H&CD) systems and the total plasma current. Open-loop data for model identification is collected during the plasma current flattop in a high-confinement scenario (H-mode). Using this data a linear state-space plasma response model for the poloidal magnetic flux profile dynamics around a reference profile is identified. The control goal is to use the H&CD systems and the plasma current to regulate the magnetic profile around a desired target profile in the presence of disturbances. The target profile is defined close enough to the reference profile used for system identification in order to stay within the range of validity of the identified model. An optimal state feedback controller with integral action is designed for this purpose. Experimental results showing the performance of the proposed controller implemented in the DIII-D tokamak are presented.

Electron paramagnetic resonance experiments on endohedral Sc3 in C82 show 22 hyperfine coupling split transitions with unusually large linewidths. Both the nuclear hyperfine coupling and the linewidths are found to be strongly temperature dependent. The data show that the three Sc ions are equivalent, and strongly suggest that they form a trimer which rapidly reorients within the C82 cage. A simple model is proposed which is in good agreement with the data.

Because spinal range of motion (ROM) is assessed routinely in clinical and research settings, a technique is needed that can be performed comfortably, quickly, and reliably. The purpose of this study was to determine if ROM data from asymptomatic subjects measured with the OSI CA 6000 Spine Motion Analyzer (OSI SMA) are reliable within and between observers. Thoracolumbar ROM, from approximately T7 to S2, was measured in all three planes in eight male and 13 female asymptomatic adult subjects (mean age = 29.7 years, SD = 5.6; mean height = 1.7 m, SD = 3.4, mean weight = 78.25 kg, SD = 34.6). A standardized protocol was used to fit each subject with appropriate hardware. Foot placement at a comfortable foot angle was standardized by the use of a template. Subjects performed three practice trials of flexion, extension, right and left sidebending, and right and left rotation. During testing, subjects performed four trials of each maximal pain-free motion. The hardware was completely removed and replaced by the same examiner, and ROM trials in all three planes were repeated. The same procedure was completed by a second examiner. Repeated measures analysis of variance and intraclass correlation coefficients (ICC [2,1] were used to analyze intra- and interobserver data. Intraobserver ICCs were 0.89 or higher for all motions. Interobserver ICCs were 0.85 or higher for all motions. Measurements of thoracolumbar ROM using the OSI SMA are sufficiently reliable within and between observers for clinical assessment and research purposes.

We describe an automated HPLC separation of the endohedral metallofullerenes such as SCC{sub 2n} and YC{sub 2n} from empty-cage fullerenes utilizing two polystyrene chromatographic columns (500 and 1000 A) in series. Rapid separation of the metallofullerene fraction from the empty-cage fullerenes (e.g., C{sub 60}) under anaerobic conditions is achieved. For the isolated SCC{sub 2n} fraction, all even-carbon-membered species from Sc{sub 2}C{sub 74} to Sc{sub 2}C{sub 104} were identified by negative-ion chemical ionization mass spectrometry. In addition, Sc{sub 3}C{sub 82} was a prominent component of this fraction. For the separated YC{sub 2n} sample, the mass spectral data indicate the presence of YC{sub 82} and all even-carbon-numbered diyttrium species from Y{sub 2}C{sub 82} to Y{sub 2}C{sub 104}. 14 refs., 7 figs.

IT was first proposed in 1985 that fullerenes can confine atoms in their interior because of their closed-cage structure. Attempts to verify this conjecture following the mass production of fullerenes have yielded metallofullerenes in bulk, and there is now good evidence that these compounds are endohedral—that is, that the metal atoms are inside. But direct confirmation in the form of structural data has been lacking, in part because of the difficulty of separating different metallofullerenes and obtaining pure crystals. Here we report the preparation of pure crystalline Sc_2@C_(84) and analyses of its structure by electron diffraction and high-resolution transmission electron microscopy. At room temperature the Sc_2@C_(84) molecules pack in a hexagonal-close-packed structure with a ratio of lattice constants c/ɑ = 1.63, the value expected for ideal-sphere packing. The molecular spacing of 11.2 A is the same as that found earlier in crystalline C_(84) (refs 5, 6). The match between our microscopic images and simulations is markedly better for endohedral models than for those in which the metal atoms reside in the lattice outside the C_(84) cages. We believe that this combination of observations points inevitably to the conclusion that the metal atoms are inside the fullerenes.

Mechanisms of protein retention in immobilized metal-affinity chromatography (IMAC) have been probed using a set of Saccharomyces cerevisiae iso-1-cytochrome c histidine variants constructed by site-directed mutagenesis. Proteins containing a single accessible histidine exhibit Langmuir-type isotherms with maximum protein binding capacities between 5 and 10% of the maximum copper loading and the capacity of the support to bind imidazole. A simple model that assumes that the copper sites are densely packed and can be blocked by protein adsorption yields binding constants for single-histidine proteins that are similar to the binding constant for free imidazole. Proteins containing multiple accessible histidines do not exhibit simple Langmuir-type behavior; they appear to interact with the support by simultaneous coordination to more than one metal ion, the result of which is to increase the apparent binding affinity by as much as a factor of 1000. The protein binding constant depends on the availability of copper sites: binding is significantly weaker at low surface concentrations of copper that presumably cannot support multiple-site interactions. The protein binding capacity drops to zero at copper loadings less than one-half the maximum, indicating that immobilized iminodiacetic acid ligands are sufficiently close together that two can coordinate a single copper ion, which precludes its interaction with a protein. Protein adsorption via multiple-site coordination has important consequences for the optimization of IMAC separations and the design of new IMAC supports.

Most pharmaceutical products are complex systems designed to meet several compendial or other performance standards simultaneously. Ideal or ‘optimum’ product composition and the manufacturing process variables are generally established after extensive experimentation. Artificial Neural Networks are pattern recognition tools that allow the development of ‘expert’ systems without having to write computer programs. With this technology it may be possible to develop formulation ‘expert’ systems to predict the formulation composition and the manufacturing process conditions necessary to achieve the desired performance standards. This report introduces the concept of a formulation expert system to predict the in vitro drug release profile from hydrophilic matrix tablets. Formulation expert systems or Computer Aided Formulation Design has the potential to reduce the time and cost of the product development process.