 |
 |
 |
|||
 | |||||
 |  |
||
 |
 |
||
 | |||
|
 |
||
| |||
|
 |
||
| |||
|
 |
||
| |||
|
 |
||
| |||
|
 |
||
| |||
|
 |
||
 |
|||
| |||
| |||
|
| David Block | |
 |
email: deblock@ucdavis.edu |
| Office: 3015 Wickson | |
| Phone: (530)754-6046 | |
| Associate Professor | ||
| Department of Chemical Engineering and Materials Science Department of Viticulture and Enology | ||
| B.S., 1986, University of Pennsylvania Ph.D., 1992, University of Minnesota | ||
Research  | ||
| ||
| Bioprocess Optimization Based On Historical Data: For all established biological processes, a wealth of data exist, whether in written form, electronic media, or in anecdotal form via process operators. Many times this data remains an under-utilized resource because efficient methods of capturing the knowledge represented by this data do not exist. By developing computational methods to examine archival data, it should be possible to use the natural and planned variations in past processing to optimize future fermentations. Though good and bad batches inevitably occur in production, reasons for these differences are not always obvious. The methods developed here will provide a way of identifying and repeating beneficial behavior while avoiding conditions that favor poor productivity or quality, thereby fully utilizing the knowledge-base for the process. Traditional experimental optimization methods such as factorial design experimentation and response surface methodology at production-scale is limited by high cost and equipment availability, as well as regulatory issues in the case of biopharmaceuticals. The first step in this process will be to develop or adapt methods for searching large databases of process information to find the information that is most likely to give the relationship between processing inputs and outputs. Clustered data that give the same information will be reduced, and problem batches with identifiable causes will also be treated differently. Once the critical information is extracted from the historical data, neural networks will be trained to establish the relationship between the chosen sets of inputs and outputs. After training of the network is complete, optimization routines will be used to find the process inputs that give the desired or optimal outputs using the neural network as a means of predicting behavior over the entire continuous space. De novo Bioprocess Synthesis: The second optimization problem on which we are working is to find an efficient means of synthesizing a bioprocess de novo. That is, this method will allow raw materials, inoculum, and processing conditions for a given bioprocess to be optimized given an objective, but no fixed initial conditions. This is significant for both the industrial fermentation and wine processing industries. One common goal of the biopharmaceutical industry is to reduce overall development time. A critical part of this development is to fix an efficient manufacturing or fermentation process, usually in a constrained time frame. Therefore, it is imperative to optimize the process rapidly and maximize the information from the small number of experiments that are possible. In a similar manner, efficient fermentation optimization is important in cases where the number of experiments is limited due to other types of constraints. Winemaking represents a case where the raw materials for experiments are only available for one short period each year. For this reason, efficient optimization will result in attaining the desired finished product in fewer years. Therefore, an evolutionary process for optimization of fermentation media (identity and concentration of nutrients), inoculum, and physical parameters such as aeration and agitation rates, pH, and temperature is being developed. The process will utilize traditional methods such as statistical experimental design and response surface methodology, along with novel means of identifying how these experiments should be planned in order to move toward the optimum in the fewest experiments. These methods will be based on results from previous experiments as well as a fundamental understanding of the underlying biology of the fermentation system. The goal of this research will be to accelerate process development by efficient management of evolving information. Biocontrol of Plant Diseases: Fungal and bacterial diseases of plants pose a serious economic problem for agriculture in California and throughout the country. These diseases are typically controlled by repeated application of chemical pesticides, or in some cases, by removal and replanting. We are collaborating with Prof. Jean VanderGheynst in Agricultural and Biological Engineering to find new biological control agents and novel means for applying them in the field. Bacteria or fungi will be isolated that naturally produce secondary metabolites which specifically inhibit growth of the plant pathogens. A process will then be developed for producing large quantities of the organism for distribution. Efficacy of the distribution methods and the anti-microbial activity will then be assessed in field trials. | ||
Laboratories | ||
| Our laboratory is equipped with four identical lab-scale fermentors. The fermentors are fully-instrumented and will be monitored and controlled by a programmable logic controller (PLC) and a supervisory control and data acquisition (SCADA) software package. The laboratory also includes an NBS incubator-shaker, incubator, and a Hewlett Packard 1100 Series HPLC. Gensym G2 software is being used on a dedicated computer for neural network and other knowledge-based analysis. Other research equipment is available in the pilot winery in the Department of Viticulture and Enology. | ||
Support | ||
| University of California New Faculty Startup Funds |
  |
Department | Faculty | Students | Directory | Alumni | Research | Admissions | Contact Us | Webmaster |
| Chemical Engineering and Materials Science Department |