What Industry Seeks from Federal Laboratories, and Examples of Collaborations
Henry T. Kohlbrand
Dow Chemical Company
We work in a rapidly changing business environment. Companies that were formed when regional economies controlled market forces are now competing in a global business environment. This has resulted in industry consolidation and changes in many business practices. This is especially true in mature industries, such as the chemical industry.
The Dow Chemical Company, formed in 1897, has had a long history of developing technology within the company as well as working with universities, national laboratories, and institutes external to the company. The Dow external technology organization today coordinates over 150 external research programs and a significantly larger number of external research agreements as well as participating in a number of business development and growth initiatives. This paper focuses on how these collaborations bring value to a multinational company, key factors that impact the success or failure of these activities, and challenges for the future. In addition, several examples of collaborations between Dow and federal laboratories are reviewed.
The Changing Nature of Business
The rapidly changing business environment in which we work requires us to continuously improve the processes that govern our activities. The move from regional to global economies has resulted in significant industry consolidation and restructuring. This has impacted both the mechanism and the role of technology development within large multinational companies. Where manufactured products were once the focus of activity, the management and exploitation of intellectual assets, once considered only an enabler to manufacturing, have now become much more economically important. In many cases, information and services have become the product.
These changes in business structure have had a significant impact on how external collaborations are managed. We are required to move quickly while maintaining a high degree of accuracy. The degree of technical understanding needs to be balanced against project timing and customer needs. Demands on product quality and performance increase while pressure to reduce prices continues. Competitive assessment has become more important, especially considering competitors that are new to
the marketplace. Joint ventures and acquisitions have become a much more common way of doing business; and purchased technology, rather than technology developed within an organization, is playing a more important role. On top of all of these changes, reengineering and restructuring have resulted in fewer employees in large companies. External collaborations must be structured to use people's time as effectively as possible.
External partnerships are now more important than ever. Understanding how to form alliances and generate value from these alliances is a critical factor in meeting the challenges of the next decade. The definition of ownership of intellectual property resulting from an external collaboration is more important than ever. In the past, the intellectual property resulting from collaboration was directed primarily toward the sponsoring company. Today it is quite common for this property to be directed toward a joint venture or later licensed or sold to another company. Often, three or more companies are involved in the use of the intellectual property. The agreements associated with ownership of this property must allow increasing flexibility in how it will be commercialized.
Overview of Dow
The Dow Chemical Company is a global science- and technology-based company that develops and manufactures a portfolio of chemicals, plastics, and agricultural products and services for customers in 168 countries around the world. With annual sales of more than $18 billion, Dow conducts its operations through 14 global businesses employing 39,000 people. The company has 123 manufacturing sites in 32 countries and supplies more than 3,500 products. In 1998, Dow divided its 14 businesses into 7 segments.
Table 7.1 shows the diverse nature of the businesses. Each of these operating segments has very different needs for different types of external collaborations. In addition to business diversity, geographic diversity also plays a role. In 1998, 40 percent of Dow's sales were from the United States, 35 percent from Europe, and 25 percent from the rest of the world.
R&D at Dow has been significantly redirected over the past few years. More effort is being placed on growth. Dow's R&D investment in maintaining current businesses is decreasing in favor of R&D focused on value growth. The R&D budget at Dow has gone from $785 million in 1997 to $807 million in 1998 to a projected $870 million in 1999. Much of this additional investment is directed toward value growth, with a significant commitment to biotechnology. We are also increasing our attention on opportunities in biomaterials, coatings and adhesives, electronic materials, and special polymer processing.
The changes in business structure at Dow have impacted the nature of our external collaborations. Prior to 1994, Dow's organization was a matrix of functions, businesses, and geographies. In 1994, a major reorganization took place. Each of the 14 businesses formed contained its own commercial, manufacturing, and R&D organization. Shared or leveraged organizations such as environmental, health, and safety; controllers; supply chain; and corporate R&D continued to exist, but their focus was directed toward the success of the 14 business units. In 1998, corporate R&D reorganized into eight capabilities. These capabilities are very aware of the business needs and are responsible for leveraging across businesses as well as developing their capability for the future.
All of the technology collaborations sponsored by Dow are supported either by a business or by a corporate R&D capability. Eighty percent of the external projects are supported by one of the 14 businesses and 20 percent of the external collaborations are supported by corporate R&D capabilities. When we establish an external technology partnership, the sponsoring business or capability will set the scope and timing of the projects. When working with Dow, external collaborators need to work with and be aware of the needs of 22 internal organizational units—the 14 businesses and the 8 corporate R&D capabilities. We have an external technology organization that addresses this interface (see Box 7.1).
TABLE 7.1 Operating Segments and Sales
1998 Sales (%)
Basic chemical products
Adhesives, sealants, and coatings
Epoxy products and intermediates
Crop production; weed, insect, and plant disease management;
industrial and commercial pest management
Hydrocarbons and energy
Fuels and crude-oil-based raw materials
Technology licensing; advanced materials for electronics; industrial biotechnology; venture capital
BOX 7.1 Corporate R&D Capabilities
External Technology Partnerships
External Technology and Intellectual Asset Management is one of the eight corporate R&D capabilities. This organization manages over 150 external agreements with over 100 different institutions. About 60 percent of these collaborations are in the United States, 20 percent in Europe, 7 percent in Asia, 7 percent in the former Soviet Union (FSU), and 6 percent in other parts of the world. The number of agreements in Europe, Asia, and the FSU is growing faster than are those in the United States. In addition, it manages 14 government contracts with a face value of $60 million.
There are many business reasons for collaborating externally for technology development and acquisition. Our external collaborations bring speed, additional expertise, and broader scope to our projects. Often, we can share the risk of working in a new area by involving partners outside of our company. The objective of our external collaborations is to complement Dow's capabilities and to fill business needs through external partnerships and to bring projects to commercialization more quickly and more cost effectively.
The major segments of this capability are shown in Box 7.2. Each of these segments represents an area of activity for the corporate capability. Cooperative research involves collaboration with universities and institutes; government-laboratory collaboration involves joint projects between government laboratories in the United States, Europe, and Asia and Dow Chemical Company. Government contract R&D includes the development of new contracts as well as the management of the $60 million in contracts currently under way. Although most of these are based in the United States, there are some in Europe and Asia. We also work collaboratively with a number of government agencies throughout the world. In the United States, examples of such projects include Vision 2020 (with the U.S. Department of Energy [DOE]), Green Chemistry (with the Environmental Protection Agency [EPA]), and the Partnership for Advancing Technology into the Housing Industry (with the U.S. Department of Housing and Urban Development [HUD]).
Our effort in geographic development includes the building of technology networks in areas of the world where Dow has a limited but growing presence in the commercial and manufacturing arenas. Currently we have efforts under way in the FSU, Israel, China, India, Korea, Australia, and Taiwan. These networks are in addition to the strong external networks in place in North America, Europe, and Japan.
BOX 7.2 Segments of External Technology and Intellectual Asset Management
The External Technology and Intellectual Asset Management capability also provides leadership and coordination for a series of technical advisory boards. These boards invite experts from outside of Dow to provide our businesses and capabilities with an external peer review of our activities as well as the opportunity to participate in collaborative projects.
The final segment of our capability is the Intellectual Asset Management Technology Center. Although the management of intellectual assets is done within each business, this technology center provides best practices for all Dow businesses in intellectual asset management, as well as providing training and career management for those involved in this activity.
The process for sponsoring external projects within Dow has changed significantly during the past ten years. At one time, about one-third of the funding of external projects came directly from corporate funding. There were virtually no projects in countries such as the FSU, China, Israel, etc. Although done with links to businesses and R&D groups, the accountability was often unclear. Today, all collaborations are sponsored by a business or corporate R&D capability, and accountability is very clear and there is clear growth in agreements in our new geographic networks.
There are several elements common to successful external partnerships. First, it is necessary to be proactive in using external partnerships. Next, it is critical to have a vision for the successful outcome of the project. This vision may be slightly different for each partner, but success should be defined from the start. A part of defining success is to outline a project where all of the partners can ''win." Assembling a successful collaboration, therefore, involves each partner understanding what success means to the other. Definitions of success are different for a company, university, and federal laboratory, but it is critical that these differences are clearly understood. Continuous communication during the project also enhances the likelihood for success. Finally, as a project goes forward, it is important to keep focused on how we can continuously improve the way that we do things. If these elements are observed, we have found that our collaborations are much more likely to succeed than if we just let them happen. One of the functions of our corporate R&D capability is to assure that these elements are a part of each new collaboration.
Federal Laboratory Interactions
Over the past 20 years, Dow has had some sort of collaboration with most of the federal laboratories. During the past five years, we have had projects with the following laboratories:
- Argonne National Laboratory,
- Armstrong Laboratory (Air Force),
- Brookhaven National Laboratory,
- Lawrence Berkeley National Laboratory,
- Los Alamos National Laboratory,
- Naval Research Laboratory,
- National Institute of Standards and Technology,
- Oak Ridge National Laboratory,
- Pacific Northwest National Laboratory, and
- Sandia National Laboratories.
There were approximately 50 agreements with these laboratories during the past five years, of which 14 are active today. The following examples represent a cross section of these agreements with a few comments on success factors for both successful and unsuccessful projects.
Argonne National Laboratory
Dow and Argonne have had a number of successful interactions over the years. One successful collaboration involved a Dow scientist who spent six months at the laboratory. He was involved in a series of technology areas and eventually focused on some coating technology of interest to Dow. This coating technology is undergoing commercial evaluation and has a good chance of being incorporated into a new product area.
Dow and Argonne were also involved in the joint development of technology related to fuel cells. Although the technology associated with the project was successful, Dow made a business decision to exit the fuel cell business and so the project was never completed.
Los Alamos National Laboratory
For several years, a Dow scientist worked at Los Alamos. Through his interactions, a number of successful projects were initiated. This scientist was chosen not only for his scientific capabilities, but also his leadership and management capabilities and his ability to develop external collaborations. By being on site for several years, he built strong relationships as well as a thorough understanding of the capabilities of both Los Alamos and Sandia.
One of the projects identified by this scientist was not successful. It involved chemical synthesis. After six to nine months, the person sponsoring the project within Dow changed jobs and the project lost its champion. The person who took his place never developed the same vision for the project and it was never completed.
Oak Ridge National Laboratory
We have had a number of successful projects with Oak Ridge in both ceramics and analytical development. These projects shared four common characteristics: They met a defined business need for Dow and met the mission of the laboratory, each had a champion in Dow and at Oak Ridge, each had defined deliverables, and there was a win-win proposition for both Oak Ridge and Dow.
Pacific Northwest National Laboratory
Our recent collaborations with Pacific Northwest have focused on information technology. This project is just beginning but exemplifies all of our current success factors. We have defined our needs and believe that there is a win-win proposition for the laboratory and Dow. Project champions in both organizations are in place and communication is continuous. This is clearly an area of technology that will not be a core technology to Dow, so it is an area in which we depend on external sources for both expertise and execution.
Sandia National Laboratories
As with Argonne, we had a project related to fuel cell technology with Sandia. Although the technology associated with the project was successful, Dow made a business decision to exit the fuel cell business and so the project was never completed. We have also had a very successful project on materials modeling. Here, Dow has a business focus, but the basic technology being developed can be
used on more than one business application. There is a clear understanding of success for the partners and strong project champions in place.
Multiphase Fluid Dynamics Consortium
Another way for companies, universities, and federal laboratories to work together is through the use of a consortium. Once a key area of interest has been identified, the consortium can be organized and the participants selected for the strengths of the capabilities that they bring to the partnership. The Multiphase Fluid Dynamics Consortium is a group of laboratories, industries, universities, and government agencies partnering to perform fundamental multiphase fluid dynamics research. In this case, it was decided to focus on solid-gas processing.
Gas-solid and gas-solid-liquid flows are identified in Technology Vision 2020: The Chemical Industry1 as critical to developing advanced chemical reactors and separations. Solids handling is a critical technology for chemical manufacture. For companies such as DuPont and Dow Chemical Company, more than 50 percent of the products sold are in a solid/particulate form. This consortium was formed to develop technology to accurately model gas-solid transport in industrial applications.
As with other worthwhile efforts, a successful consortium activity does not develop overnight. Prior to 1995, a series of collaborations were taking place across the chemical industry. Dow, DuPont, Chevron, and Exxon all had projects under way at a number of universities and federal laboratories. The CRAY Chemical Process Computational Fluid Dynamics (CFD) Users Group, The Reactive Multiphase Flow Workshop held at Los Alamos in May 1995, the First Engineering Foundation Conference on CFD in Chemical Reaction Engineering held in San Diego in October 1996, and the DOE Office of Industrial Technologies (OIT) Roadmap Workshop on CFD held at the University of Maryland in October 1996 all demonstrated that a significant collaborative effort was required to address the opportunities identified in the Vision 2020 document.
In 1998 the consortium was formed. The OIT supported two related projects in the area of CFD, which form the keystones of the consortium. DOE funding was awarded at a level of $1 million per year. The industrial leaders (Chevron, Dow Chemical, Dow Coming, DuPont, Exxon, and Siemens Westinghouse Power, with computer maker Silicon Graphics and engineering software companies AEA Technology and Fluent) provide leadership and cost sharing. DOE funding supports work at Lawrence Berkeley, Los Alamos, Oak Ridge, Pacific Northwest, and Sandia National laboratories and the Federal Energy Technology Center. The consortium also provides funding for related research projects at Clarkson University, Illinois Institute of Technology, Princeton University, Purdue University, and Washington University-St. Louis.
Each partner in the consortium contributes through their strengths and knowledge in a variety of areas. Their knowledge in defining needs, experimentation, equipment design and construction, and modeling as well as application experience all come together to guide the projects. A clear definition of industry needs, project milestones and timetables, and desired outcomes help to measure progress and direct activities. A path to share the results through a series of meetings, publications, and presentations is also in place. Also, strong project champions in each company, university, and federal laboratory play a key role in making sure that the project targets are met. Tyler Thompson of Dow and Brian Volintine of OIT played key oversight roles in organizing the consortium.
Several key factors have contributed to the success of this effort. At the beginning, a lot of effort went into establishing a shared vision among all of the participants. The team leading the effort is both self-organized and self-directed. Each organization participating is committed to a portion of the project, and the industry is ready to implement the results as they become available. It is also important to note that the technology resulting from these projects results in precompetitive tool development rather than invention and discovery that might be considered proprietary by the participants.
All institutions are proud of their capabilities. This is true of industry, federal laboratories, universities, and institutes. Each institution has a few capabilities that truly differentiate them from others. When we start to form a relationship with a new institution, we try to identify these differentiated areas to begin projects in these areas. It is very important that federal laboratories, universities, and institutes clearly communicate what they consider their strengths to be and make it easier for collaborators to find their way through "organizational mazes."
What We Have Learned
In our work in developing external technology partnerships, we have found a series of elements that need to be present in order to have a successful project. These are listed in Box 7.3.
A project that is well defined at the beginning has a much better chance of success than one in which the definition is vague. An understanding of the company's needs as well as the laboratory's capabilities sets the foundation for defining the project. In our case, each project needs to fit a defined need by one of our business units or one of our corporate capabilities. A project without such a connection has little chance of being funded. Early in the project, the agreement on results, how to measure them, and intellectual property ownership contributes to the focus of the principal investigators and leaves less room for misunderstanding between partners as the project progresses. Effective communication during the project helps to keep the project on track and allows the partners to agree on necessary changes in plans. Finally, successful projects need to deliver results.
BOX 7.3 Elements for Successful Projects
Most successful interactions involved Dow scientists and external technology staff working with federal laboratories to clearly understand their capabilities before any projects were defined. Almost no successful interactions occurred as a result of laboratory outreach visits (although once a relationship is established, visits from laboratory staff and scientists help to build understanding among company scientists). Each successful project had a project champion inside of Dow and a project champion at the federal laboratory. Where laboratories made a special effort to identify their areas of strength, interactions seemed to proceed more quickly and more effectively. Some projects had unique challenges because some of the company participants were outside of the United States. The need to involve international teams is increasing as the global presence of companies continues to grow.
In building partnerships, it is important to remember that there are four partners in each company federal laboratory collaboration:
- Federal government,
- Federal laboratory,
- Sponsoring business or research capability in the company, and
- The company.
Many times an agreement takes additional time to finalize because one of the four partners is not taken into account. It is important to recognize that the needs of the federal government and federal laboratory do not always align, and, similarly, the needs of the company and sponsoring business unit many not always align. Reconciling these issues early in developing a relationship can avoid misunderstandings later in the project.
Flexibility is increasingly important. In the examples of Dow and federal laboratory interactions, it was pointed out that a project can be technically successful, but may not be completed because the business needs of the company change. Needs of both organizations may change over the life of the collaboration. Continuous communication helps to manage these situations more effectively.
The exchange of people facilitates knowledge and technology transfer. It would be ideal to have company scientists present at each federal laboratory, university, and institute. When company scientists have been stationed at a federal laboratory, the number of quality collaborations has greatly increased. Similarly, when federal laboratory scientists have worked at an industrial facility, the number of quality interactions has also increased. Unfortunately, this is not practical. The number of people required and costs associated with this effort would be prohibitive. This is possible in a few, very select cases, but alternatives to this mechanism for knowledge and technology transfer must be found.
Any industrial-federal laboratory collaboration must be cost effective and cost competitive with alternatives. This is an important aspect of defining success at the beginning of an interaction. Part of this discussion involves the ownership of intellectual property resulting from the joint activity. This tends to be easy for precompetitive projects, but can be increasingly difficult for projects that are closer to market implementation. In some cases, ownership of patents resulting from joint efforts may be governed by legislation and may restrict the scope of the project.
Finally, people make the difference. With the correct people involved and teamwork established, the likelihood of project success can be increased.
There are always opportunities to improve the ways that we work together. It is often difficult for the federal laboratories to know where they should be developing expertise for the future. Better
communication by industry of its needs is an area in which we need more focus. Efforts such as Vision 2020 have begun to collect some of this information, but in general, more specifics are needed.
We should start with an understanding of how to make collaboration a win-win situation. Key elements need to be defined from the beginning. Understanding how sensitive or proprietary the project area is to the company, or how important it is for the federal laboratory to publicize the collaboration, can shape the agreement more effectively. In one interaction we had with a federal laboratory, the premature publication of the collaboration by the laboratory made the project impossible to complete.
Intellectual property issues need to be addressed more effectively. In the past, collaborations typically involved a company and a federal laboratory. Simple assignments of the patents or provisions for the company to license the technology were adequate. The DOE cooperative research and development agreement (CRADA) "Subject Invention" definition can confuse intellectual property ownership. Today, the collaborations may involve several partners, the results may be used by joint ventures, or even more-creative business structures may need access to the intellectual property developed by the collaboration. Increased flexibility on how this property is managed is a key issue for the future.
When developing relationships with federal laboratories, it is sometimes desirable for the company to invite federal laboratory employees to act as consultants. For scientists who are employees of the U.S. government, U.S. law declares their inventions to be U.S. property (42 U.S.C., Sec. 5908) and makes it difficult for them to work as consultants. Some privately operated federal laboratories apply these same rules to their employees. Others operate on a different set of rules. This is an area in which improvements and clarity would be welcomed.
Industry looks for excellence, partnership, cost effectiveness, and versatility in external collaborations. The federal laboratories are a great resource for joint technology development. A better understanding of the laboratories' strengths by industry will aid in the most effective utilization of their capabilities. Key success factors include defining a project in which all partners' needs are met, having a project champion in each organization to facilitate progress, communicating continuously, and defining flexibility into the agreement so that it can change if necessary. A moderate effort to broaden options for ownership of intellectual property and involvement of federal laboratory employees as consultants could facilitate additional collaborations with industry.
Bruce Harrer, Pacific Northwest National Laboratory: Most of the employees of the national laboratories are not federal employees, so I do not understand the issue over intellectual property.
Henry Kohlbrand: I cannot answer that from a legal point of view. Intellectual property covers federal employees; it covers apparently a broader spectrum than that, according to our lawyers. This was a problem at Sandia. Sandia lawyers said that they couldn't sign this kind of a secrecy agreement because we're covered under it. So I can't answer the legal question of that, but I can tell you that this question has been raised by the legal staffs of several of the federal labs that do not have federal employees.
William Millman, U.S. Department of Energy: The difference between federal laboratories and national laboratories is that federal laboratories have federal employees. They are government owned and government operated. The DOE's large multipurpose national laboratories are government owned,
but contractor operated. The result is that laboratory employees are employees of the contractor, such as the University of California or Lockheed Martin.
Henry Kohlbrand: I cannot adequately answer the question. It's a question that we should highlight from this discussion. I know that the lawyers are worried about it.
Ashok Dhingra, Dupont: Would you comment on Dow's collaborations in Asia Pacific, specifically how you deal with the communication and proprietary issues?
Henry Kohlbrand: We have had some good luck. We've had an office of science and technology in Japan for a long time, and we found that working in Japan is a little different from a cultural point of view. Japanese agreements are wonderfully short. The Japanese say that we are going to work for mutual benefit and that the details can be worked out later. If you are not comfortable signing something like that, then you're in trouble.
We have several interactions going on in China and India. Because of the intellectual property issue, we have tended to stay away from areas of technology that are what I call sensitive within the company right now. So we might be working with the National Chemical Laboratory in Pune or we might be working with the Indian Institute of Chemical Technology on some specific projects of developing some new chemistry or some new catalysts. In China we are doing a lot of synthetic organic chemistry with some of the Chinese institutes, and we are doing some engineering work with one of the large universities in China.
So we try and recognize that there are differences in the way proprietary information is handled within these countries, so we don't bring a sensitive project to those countries, but are effectively working with them on other things.
Christopher Hill, George Mason University: My comment is on the definition of a federal lab and a national lab. Some of the federal labs that belong to DOE have been designated "national labs," but all of them are federal labs.
As the structure and decision making for collaborations within Dow have changed in recent years, has there been a concomitant change in either the functional level or responsibility of the individual within Dow who has the sign-off authority on engaging in a collaboration? Has that moved up or down in the organization or away from the R&D management? Where does it reside?
Henry Kohlbrand: Depending again on the size of the agreement, I would say, in general, that the signing authority may have moved down a notch because we have gone from roughly 13 levels of management down to about 5. We cut out extremely large numbers of management layers. But for large programs, we need to get business approval. So it is not just the R&D community that approves collaborations.
Sometimes we have to help the business people understand why it's necessary to be spending this kind of money outside the company and what we expect to get for it. In some cases if it's money that's being given as a nonrestricted grant to avoid overhead and to do some other things with the university, then the business people have some struggles with that. So we have to do additional communication and additional work within the organization. For most of the research projects that are taking place, approximately 80 to 90 percent of them, they are still handled within the R&D umbrella, and we don't have those types of problems.
Hank Whalen, PQ Corporation: Are CRADAs increasing or decreasing with Dow?
Henry Kohlbrand: Right now we may have one or two more CRADAs this year than we had a couple years ago. I don't know if that is a substantial increase. Again, CRADAs are a good way to interact with a laboratory as long as you work together on how the outcome of the research is going to be done. We are seeing more involvement in the Advanced Technology Program (ATP). We are more interested in the company and ATP-type programs than we have been in the past because we can get closer to products with ATP projects than we can with CRADAs. So I would say that it is flat right now. I am unsure if there is a reason to drive it up and down at this point.
James Seiber, U.S. Department of Agriculture: You gave a number of examples of cooperation with national laboratories. What are your experiences with other federal organizations, such as the Department of Agriculture, the National Institute of Standards and Technology (NIST) and the EPA? And maybe more importantly, how do such organizations make the first contact within your company?
Henry Kohlbrand: I am unaware of any interaction that Dow has had with the Department of Agriculture. That doesn't mean that there aren't any because Dow is a big company. In terms of the other organizations, we have worked extensively over the years with the Department of Transportation because we transport a lot of stuff. We have also worked with the Coast Guard and the Department of Commerce. We have a lot of interactions with NIST because NIST is a marvelous place to do some types of work, so that is a continuing effort. We have a person who goes out two or three times a month to foster some of these interactions. When working with the EPA, we have had both good and bad experiences, as you might well imagine, because there are a couple different parts to the EPA. But I had some good interactions with the technical side of EPA, when we were collaborating on incineration when I was working in that area about ten years ago. So Dow has a lot of activity going on with the agencies, and we have two people in Washington, D.C., to facilitate these activities.