Industrial Research Laboratories
Where does technological innovation happen? We
have talked about how engineering schools struggled to put science and
engineering together. The same struggle had to be worked out in
As science became more significant inventors like
Thomas Edison who didn't have formal education found themselves at a
- Who comes up with new technology?
- How are those people trained?
- Where do they work?
Royal Gunpowder Mill, Ireland
Du Pont: (
DuPont Company History , Hagley Museum )
- E. I. du Pont started a gunpowder business
in 1802. There wasn't much innovation, though Du Pont sometimes
took samples of saltpeter
to a Philadelphia chemist for analysis before accepting the shipment.
du Pont was the first leader of the family
business to go out and investigate what science could contribute--he
a degree in chemistry at the University of Pennsylvania and made some
significant innovations, such as a way to substitute sodium nitrate for
potassium nitrate. Organic chemistry of dyes was turning out to
be related to explosives--in 1880 Lammot started his own company, the Repauno
Chemical Company, to make dynamite. But he was killed in an
Inventor of Dynamite
- Lammot du Pont's role as
taken over by Charles T. Jackson. Hired chemists for
control, investigation of complaints, evaluation of new inventions, and
improvements in the factory process, but not specifically to develop
- The invention of smokeless powder
threatened DuPont--they saw the power of patented innovations. In
1902 the laboratory was separated
from production and assigned to develop new products, though they often
up doing further development on patents bought from Germany and they
did some testing and manufacturing control.
- By 1912 the Chemical Department had 120
and at the end of World War I new product research took of to find a
for idle gunpowder factories. Branched out into plastics,
and paints; rayon
1920, synthetic ammonia 1924, cellophane 1927 (all purchased
French technology). For example, they bought the patent for
in 1924, and by 1927 had developed a moisture proof cellophane.
But they found it hard to establish leadership--they invested some $40
million in dyestuff research after 1917 but found that they mostly
ended up producing colors that had been pioneered in Germany.
- The new products that finally came entirely
inside the lab were fibers--neoprene (synthetic rubber) in 1930 and
Edison's Menlo Park Laboratory (
Henry Ford Museum
General Electric: (
GE Research Lab History )
- Edison claimed he didn't need science, but
have almost an industrial research lab where he hired inventors and
mathematicians to work out the details of his ideas.
- General Electric started to look towards
more systematic research as Edison's patents expired--the basic one in
1894. GE's carbon
filament faced competition from Nernst's ceramic filament, which
had the American patent on. The company was following a strategy
buying patents from inventors, but that was costing a lot.
- In 1900 General Electric's chief consulting
engineer, Charles Steinmetz, proposed the establishment of a laboratory
for original research entirely separate from the factory and immediate
production problems. "We all agreed it was to be a real
scientific laboratory." Modeled on
experiments already underway in Germany to bring scientists into
- Headed by Willis
R. Whitney, a professor of physical
chemistry at MIT with a Ph.D. from Leipzig (under Wilhelm
Whitney was tired of low salaries, slow promotions, and lack of
and interested in industrial problems. GE was a disappointment at
first--he found himself temporarily housed in a barn at GE because of
space problems. He solved that problem by burning down the barn.
- Lab went from 8 people in 1901 to 102 in
to 40% of the staff had scientific training. Whitney believed
the way to develop new products was to do exploratory scientific
was willing to buy that idea because Whitney kept clear that the goal
new products. University gave social prestige, research freedom,
and professionalism. Lab offered money and time to work on
research without demands for teaching and theory.
- Whitney's staff got to work on light bulbs
found a way to improve the carbon bulb 25% by baking the filament.
- But all that was rendered obsolete by the
in Germany of more efficient lamps with filaments made of rare metals
such as tungsten and tantalum (by Carl Auer von Welsbach, Walther
Nernst and Werner
von Bolton). In 1906 GE had to buy patents at a cost of
$300,000--a great defeat for the lab. William Coolidge said:
"These were the expenses
that the Laboratory had been founded with the purpose of
preventing." Even then there were tremendous problems with
putting the new technology into
use, and the laboratory worked on those without much success (see
T. P. Hughes American Genesis p. 167).
a breakdown, partly mental and partly untreated appendicitis. He
returned to the lab as a manager, not as a researcher.
- His replacement as leader of research was William
D. Coolidge, a Ph.D. from Leipzig, also a physical chemist.
hired in 1905 at a salary of $2400--50% more than he was making at
His 1913 patent on a process for making an improved
tungsten filament saved
the lab. But this was still an improvement of an existing
not something radically new.
Langmuir was the first to break that barrier.
Had a Ph.D. from Gottingen where he had studied under Walther Nernst
of the 3rd law of thermodynamics) and a commitment to pure science, but
had little time for research at his job at Stevens Tech.
to GE by the idea of spending full time on research and by better
What were the limits? Company owned patents, required reports of
and daily notebook. Langmuir studied chemical reactions a low
pressures (inside a light bulb) and published a stream of papers in
physics and chemistry (averaged 5 papers a year from 1912 on) and also
a steady stream of patents.
- In 1916 Langmuir invented the gas-filled bulb--gave
GE market domination (96% of American incandescent-lamp sales in
1928). He won the Nobel prize
for his work in surface chemical reactions (the work
that had led to the lamp) in 1932.
- The lab had proved itself at GE, but the
was that to get radically new products or new approaches you had to let
scientists do fairly open-ended scientific research, rather than
telling them what to
work on. Scientists had to prove themselves first, but then the
scientists were allowed to pick their own research projects.
Industrial research allows corporations to control
- Key features of industrial research lab:
- freedom from operational
employees have to solve factory problems they won't get to researching
- hiring Ph.D. scientists--because a Ph.D.
you original research
- allowing scientific publication--because
a reward system scientists care about
- defensive and offensive patenting--patent
your competitors need, not just things you need
- Attracted scientists who were inventive and
practical-minded but had no taste for the entrepreneurial activity and
the financial risk taking
in which the independent inventors had had to involve themselves.
this page written and copyright
Pamela E. Mack
last updated 10/10/2005