Spring 2005 Projects

Summaries of each project are below. Projects span a broad range of technologies:

• Ice-cream production
• Titanium production
• Nanomanipulators
• Database management systems
• Portable spectrometers

1. A BETTER WAY TO MAKE ICE CREAM: This cheaper, more energy-efficient way to process frozen foods could change large-scale ice-cream production as we know it—or could even create a whole new creamy dessert.

The $17.9 billion worth of ice cream produced in the U.S. each year is made via a process that hasn’t changed much since the invention of the hand-cranked freezer in 1846. Prof. John Brisson's research team proposes a process that could substantially change the way ice cream is made and improve the end product. It involves making ice cream using liquid carbon dioxide. The process uses less power than current ice-cream production methods and would reduce capital investment and maintenance costs. Additionally, the new freezing method could even lead to new products, including unusual, complex, molded desserts and reduced fat ice cream with the creaminess of regular ice cream. The process will likely scale well and could potentially be used in point-of-sale production, as well as large-scale manufacturing.

The frozen-dessert market is highly fragmented with no company accounting for more than 10 percent of the total market share. The i-Team‘s task will be to identify the best commercial prospects and target markets.

2. CLEANER, COST-COMPETITIVE TITANIUM PRODUCTION: Prof. Donald Sadoway's lab has developed an innovation that reduces the cost of producing titanium, which could open up previously unattainable markets for this remarkable, corrosion-resistant metal.

The innovation is a new process for the production of titanium metal: the electrolytic extraction of titanium from titanium oxide. It avoids the use of chlorine and carbon, both of which are present in all commercial processes and contribute to the high cost of production by requiring a large number of steps (unit operations) and considerable effort to contain toxic emissions. The new process must overcome some technical obstacles, but its development would dramatically reduce the cost of titanium and hence result in greater utilization of this remarkable metal. There are huge markets in aerospace, marine structures, and, ultimately, automotive applications.

An i-Team could help provide a rigorous assessment of market potential as a function of metal price and cast this in the light of a proper evaluation of cost of production by the proposed new process.

3. HEXFLEX NANOMANIPULATOR: Current developments in nanotechnology are limited by the difficulty of manipulating objects to extreme precisions. This invention by Prof. Martin Culpepper is an elegant and inexpensive solution to the problem.

Nanomanipulators are the "construction equipment" used to move, align and inspect nano-scale products. State-of-the-art robotic nanomanipulators are the size of a bread box, cost more than an SUV, and struggle to move/align/assemble products on the nano-scale.

The HexFlex Nanomanipulator incorporates recent advances in precision engineering and robot design to easily achieve nanometer-level accuracy and 300X better position stability at 3X lower cost. In our Deshpande-funded research, we are working to integrate technologies which enable: (1) better performance - make the HexFlex move faster and provide more accurate performance, thereby enabling manufacturers to make better products and make them faster; and (2) reconfigurability: the novel robot design may be "tweaked" so that it performs the tasks of several robots. This may enable us to design/build/sell one machine which can do the job of several machines (perhaps from different markets). We have identified three candidate applications:

1. Wafer inspection for nano/micro-manufacturing
2. Micro-photonics manufacturing and R&D
3. Molecule/DNA manipulation

Prof. Culpepper's lab is well equipped to complete the technical components of the work. The i-Team will work with the lab to better understand the potential of each market, determine which market is best to pursue, and determine if it is feasible to address several markets with a reconfigurable HexFlex. This collaboration will form the seed for a team which will subsequently work to commercialize the HexFlex within the next year.

4. COLUMN STORE - A NEW DBMS ARCHITECTURE: Commercial database management systems are designed for update-intensive applications, leaving an opportunity for Prof. Michael Stonebraker's radical new hybrid approach that can address the data warehouse market.

Wal-Mart has 460 terabytes of data stored... at its Bentonville headquarters. To put that in perspective, the Internet has less than half as much data, according to experts. That much information results in some interesting data-mining. Did you know hurricanes increase strawberry Pop Tarts sales seven-fold? - The New York Times

All major commercial database management systems (DBMS) store data with a row (or record) orientation. This has proved very successful in transaction processing applications, but this architecture is very slow for data warehouse applications. Warehouses are used by Fortune 500 companies to predict purchasing behavior and manage their complex supply chains. In this environment, data from transaction systems is loaded periodically into a historical store, and analysts run ad-hoc queries to garner intelligence about the business and hope that the query will be complete by the next day. The proposed read-optimized hybrid approach would provide performance an order of magnitude faster than the traditional row store DBMS.

The purpose of this project is to prototype a column-store architecture. The i-Team's challenge will be to better understand the customer requirements and best path to market for this technology.

5. SLIM SPECTROMETER: Prof. Vladimir Bulovic's compact, rugged, inexpensive, optical spectrometer would greatly improve the performance of field applications like point-of-care medical devices, food analyzers, color sensors, field-installed chemical and environmental sensors, and personal environmental monitors, or even create new breakthrough applications.

Today's most compact spectrometers use silicon photodetector arrays capped with expensive interference bandpass filters that make these devices commercially impractical. Most of them are as big as portable computers and cost thousands of dollars. The slim spectrometer alleviates the need for expensive optical components (e.g. lenses, gratings) and intricate assembly during manufacturing. At the same time, the vastly reduced number of components and their complexity enhances the ruggedness of the device. The slim spectrometer would enable devices the size of handheld PDAs and costing hundreds of dollars.

The challenge for the i-team is to 1) consider the business opportunities in applying slim-format spectrographs to conventional spectroscopic applications and 2) envision breakthrough applications enabled by the small format of the spectrometer. What new markets would you enable/create? For example, what if we could inexpensively monitor anthrax or other biochemicals in the air by attaching the spectrograph to every traffic light? Wouldn't the government desire such a device in every city intersection?