Imaging the Embryonic Universe – Rome, 23.11.2006 Forum

Italy - USA

Paolo de Bernardis and Andrew Lange

2006 Balzan Prize for Observational Astronomy and Astrophysics

For their contributions to cosmology, in particular the BOOMERanG Antarctic balloon experiment.

For the past 15 years, we have been fortunate to work together with each other and with a remarkable group of students, postdoctoral fellows, colleagues, and mentors to attempt to answer some of humankind’s oldest questions.  How did the Universe begin?  What is it made of?  What is its ultimate fate?  It has been an exciting adventure both intellectually and, not infrequently, physically and emotionally. Our story has three beginnings and, as yet, no end.

First Beginning:  The Big Bang

The Universe began 14 billion years ago in a violent explosion that cosmologists refer to, wryly, as the “big bang”.  Three pillars of evidence support this view: the ongoing expansion of the Universe; the relative abundance of light elements synthesized by nuclear fusion during the first few minutes; and, most dramatically and directly, the afterglow of the primeval fireball, visible today as an almost perfectly uniform flux of microwave radiation that fills the entire sky. It is this Cosmic Microwave Background (CMB) that we have devoted our careers to studying, and that has played so large a role in the cosmological discoveries of the past decade.

The closest parallel between the biblical account of creation and our modern scientific understanding is the phrase “let there be light”.  At the extremely high temperatures present at the big bang, matter and anti-matter were spontaneously produced in copious abundance.  As the Universe expanded, however, it rapidly cooled. Within seconds, matter and anti-matter annihilated to produce gamma-ray photons.  One minute after the big bang, virtually all of the energy in the Universe was in the form of this intense light.  The only matter that remained was a tiny residue – roughly 1 nucleon for every billion photons – due to correspondingly tiny asymmetry in the initial production of matter over anti-matter.  This ratio remains unchanged today.  The tiny residue of matter is what we and the stars are made of.  For every nucleon in the Universe today, there are roughly 1 billion CMB photons streaming through space.  Remarkably, though the CMB photons have steadily lost energy with the expansion of the Universe, the energy density of the CMB still dwarfs the energy density of all other light emitted by all of the stars in the Universe.  They are old and cold (currently 2.7 Kelvin), but they win by sheer numbers.

For the first 380,000 years after the big bang, little changed other than the continuing expansion and cooling.  The Universe during this embryonic phase is simple: an opaque and almost perfectly homogenous plasma of light and ionized matter.  The dynamics are easily calculated. Regions in causal contact, or “within the causal horizon” with slightly higher matter density begin to collapse gravitationally, but “bounce” due to the pressure of the photons that are dragged along with the ionized matter.  No structure can grow as long as the matter remains ionized.  Acoustic waves of density fluctuations thus race through the Universe creating a harmonic series of characteristic spatial scales, related to the size of the causal horizon, that evolves as a function of time.

After 380,000 years, the plasma cools to ~ 3000 K, precipitating a dramatic change.  The energy of the photons is now small enough that the nucleons and electrons can combine to form atoms. The fog clears. What had been opaque plasma becomes perfectly transparent hydrogen and helium, with trace amounts of other light elements formed in the first few minutes.  To very good approximation, the photons never again interact with matter.  Free streaming through space, following the geodesics that Einstein taught us define the curvature of the Universe, they continue to lose energy in proportion to the expansion of the Universe, till – some 14 billion years later – a tiny fraction of them are captured in our telescope.

If we look at the sky with a properly designed microwave telescope, then, we have the opportunity to directly image the embryonic Universe.  The term “embryonic” is appropriate in several respects.  First, if we create an analogy between the age of the Universe and the age of a human adult, then imaging the Universe at age 380,000 years is equivalent to imaging the human just hours after conception.  Like the human embryo at this stage, the images of the Universe that we will obtain will look nothing like the adult form; the first atoms have only just formed, and there are only the faintest hints of the density fluctuations that will later grow to become clusters, galaxies, stars and planets.  Yet, just as the biologist can today readily decode the DNA of an embryonic cell and understand what it will eventually become, so the cosmologist can study the information imprinted in the CMB and, comparing with the “adult” form of the Universe that we see today, understand the geometry of the Universe, its content, and the laws governing its growth from embryo to adult.

Second Beginning:  Discovery and Characterization of the CMB

The CMB is not faint – at wavelengths of several millimeters it outshines even the emission from our own galaxy.  Yet it went unnoticed by radio astronomers for many years, precisely because it is so uniform. It was not till 1963 that Arno Penzias and Robert Wilson, in the midst of a painstakingly careful calibration of a Bell Laboratories communications antenna, discovered that the entire sky was aglow in microwaves with a brightness corresponding to a temperature of about 3K.  This discovery, for which Penzias and Wilson received the 1978 Nobel Prize in physics, immediately elevated the big bang paradigm from one of several to the standard cosmological model.  It also stimulated experimental efforts all over the world to image this newfound relic of the big bang.

For the first 29 years these efforts were a source of both frustration and wonder. Early predictions of the amplitude of temperature differences that would be seen from one part of the sky and another were proved wrong as physicists mapped the sky with increasingly sophisticated instruments and set increasingly precise limits on any deviation from perfect uniformity. By 1990, despite dozens of experiments using a variety of highly specialized, state-of-the-art microwave receivers deployed on mountaintops, carried aloft on balloons telescopes and even placed at the South Pole, there was still no hint of structure on any angular scale. 

The extreme uniformity of the CMB posed a fundamental problem. In the original big bang paradigm, there was no way to explain why the temperature of the CMB should be perfectly uniform across the sky.  This is because regions separated by more than a few degrees on the sky would never have been in causal contact with one another, and thus had no way to equilibrate to the same temperature.  In 1980, motivated by this and other problems, Alan Guth proposed a new and startling revision to the big bang called inflation.  In brief, Guth proposed that the entire observable Universe had sprung from a volume smaller than the nucleus of an atom.  Prior to inflation, the tiny dimensions allowed the Universe to rapidly equilibrate to a single temperature. Then through some quirk of physics at very high energies, the Universe “inflated” expanding explosively at superluminal velocity from atomic to macroscopic dimensions. This process occurred in a very small fraction of a second and then turned off, allowing the Universe to continue to expand and cool at a more modest rate.

Though the idea was fantastic, it neatly explained the isotropy of the CMB and solved other problems with the conventional big bang.  With important contributions from Slava Mukhanov, Andrei Linde, Paul Steinhardt, and many others, “inflation” became the standard cosmological paradigm, even though it relied on extraordinary and poorly understood physics occurring at energies well beyond the realm ever probed in terrestrial laboratories. It became urgent to find a way to test the inflation paradigm.  The CMB would eventually provide this test.

Our understanding of the CMB took a great leap forward in the early 1990’s with the first results from NASA’s Cosmic Background Explorer (COBE).  COBE carried two CMB experiments.  A team led by John Mather used a novel spectrometer to search for deviations from the perfect “blackbody” spectrum that the CMB would have if it was, in fact, uncontaminated by interactions with matter in the course of its long journey. George Smoot led a team that used microwave radiometers to search for any hint of spatial structure in the CMB on angular scales of 7 degrees to 90 degrees.  Both results were spectacular.  Mather’s team showed in 1990 that the spectrum was exactly the blackbody spectrum first derived by Planck almost a century ago, with unprecedented precision. In 1992, Smoot’s team announced the first hints of structure, at the level of ~ 10 parts per million, and with an angular distribution consistent with the expectations of inflation theory.  A new phase of CMB research was born.  For these accomplishments, Mather and Smoot were awarded the 2006 Nobel Prize in physics, just one month ago.

The impact of the COBE results was great.  Though several experimental groups around the world, including each of ours, had understood the enormous potential of the CMB for some time and had been making steady progress in developing instrumentation, the COBE results convinced a larger fraction of the physics community that these measurements were both important and worth investing in.  Funding for new experiments flowed more freely, and the experiments attracted the very best students. 

Theoretically, detailed predictions were made of the nature of the anisotropies that would be revealed by experiments with higher angular resolution than COBE. Insightful analysis and increases in computer speed combined to allow detailed simulations to be made of how the detailed form of the temperature fluctuations would depend on the geometry, content, and dynamics of the Universe.  A methodology was worked out that would allow the basic cosmological parameters to be determined from the “power spectrum” of temperature fluctations – the amplitude of temperature fluctuations as a function of angular scale. 

As a result of these advances, it became clear that the CMB would provide a dramatic test of the inflation paradigm. Inflation predicted (i) that the initial density fluctuations had equal amplitudes on all length scales (ii) that the fluctuations were adiabatic and (iii) that the geometry of the universe was exactly flat.  Each of these manifested itself as a specific feature in the spectrum of the CMB temperature fluctuations. The COBE results were consistent with (i) but did not probe it very precisely, and did not probe (ii) or (iii) at all. To put inflation to the test, one needed to resolve the characteristic angular scale of the fluctuations which, if inflation was correct, would be approximately 1 degree.

Third Beginning(s):  Mentorship and the Birth of BOOMERanG

Even before the COBE results, it was clear that we would need new technologies if we were ever to image the structure in the CMB with high fidelity.  The COBE results bore this fact out:  it took years of integration in orbit to tease out hints of structure that was unresolved and barely visible in the noise.

During the 1980’s, groups led by Paul Richards in Berkeley and Francesco Melchiorri in Italy began to apply a new technique to anisotropy measurements.  Unlike the microwave amplifiers traditionally used by radio astronomers and used on COBE, they used an approach – direct bolometric detection – that at first glance appears crude. A bolometer is a general term applied to any detection method that measures the temperature rise when the energy that is to be detected is absorbed.  Applied to measurements of the CMB, a surface engineered to absorb microwaves is placed in the focal plane of the telescope and coupled to a very sensitive thermometer.  

If the bolometer is cooled to temperatures well below the temperature of the CMB – typically a few tenths of a degree above absolute zero – the device can be made so sensitive that it becomes an ideal detector, limited only by the statistical fluctuations intrinsic to the incoming signal. The problem is that the bolometer is an indiscriminate detector of everything. Early attempts to use these devices were often frustrated by responses to spurious signals. The bolometer showed great promise, but needed to be tamed if it was ever to be useful.

In 1980, Lange was completing his undergraduate thesis at Princeton with David Wilkinson, one of the great pioneers of CMB research.  Wilkinson saw the potential impact that bolometers might have on the field.  On his suggestion, Lange went to Berkeley to work with Paul Richards, who had long applied bolometric technology to many fields of physics, and most recently to measurements of the spectrum of the CMB with his students John Mather and David Woody.  Working first in Richards’ lab and then in his own upon joining the Berkeley faculty in 1987, Lange set to work adapting bolometric detectors to CMB observations.  Over the course of a decade, in addition to improving the intrinsic sensitivity of the detectors to the point that they became virtually ideal detectors, he and his students developed techniques to eliminate low frequency noise, to reduce sensitivity to spurious microphonic and electromagnetic interference, to reduce response to cosmic rays, and to optimally couple the bolometer to the telescope optics, as well as cryogenic techniques to more conveniently cool the detectors to sub-Kelvin temperatures. These innovations found application in a variety of CMB experiments, most notably MAX, a balloon-borne telescope that targeted angular scales of 1 degree, where it was thought the signal to be detected would be largest. The MAX telescope, a collaboration between groups led by Lange and Richards at Berkeley and Phil Lubin at Santa Barbara, was flown over the U.S. desert 5 times between 1989 and 1994, and repeatedly found evidence for temperature fluctuations on the 1 degree scale that it was tuned to. [Meinhold et al. 1993, Gundersen et al. 1993, Devlin et al. 1994, Clapp et al. 1994, Tanaka et al. 1996].

            In 1981 de Bernardis was working at his laurea thesis in the group of Francesco Melchiorri, another pioneer of CMB research. Melchiorri had used stratospheric balloons to fly bolometric CMB anisotropy experiments, measuring the large and intermediate-scale anisotropy of the CMB. Facing the challenge of measuring photon noise fluctuations in the CMB, de Bernardis was introduced to the use of cryogenic bolometers and of the related experimental techniques. He obtained a researcher position at University La Sapienza in 1984, and started to develop a real multi-band telescope to get high resolution data of the CMB. He compared different measurement techniques with instruments working from Antarctica [Dall’ Oglio and de Bernardis, 1988] and from balloon-payloads. He finally set up an Italian collaboration with the cryogenic group of Franco Scaramuzzi in Frascati and the attitude control system group of Andrea Boscaleri in Firenze. With the group in La Sapienza in charge of telescope, detectors, electronics, calibration and data analysis, they built a 1.2 m balloon-borne telescope, ARGO, which was flown for the first time in 1988 [de Bernardis et al. 1990]. ARGO detected degree-scale anisotropy of the CMB in its 1993 flight [de Bernardis et al. 1994], and set important limits to the level of Galactic contamination in mm-wave CMB measurements [Masi et al. 1996].

The Birth of BOOMERanG

In 1991, Richards spent part of a sabbatical in Melchiorri’s lab in Rome. NASA had recently demonstrated the capability of keeping a balloon-borne payload aloft for over 10 days by launching from the edge of the Antarctic continent at the peak of the austral summer, when the high-altitude winds carried the payload around the South Pole at roughly constant latitude.  Melchiorri mentioned to Richards that within Italy such an experiment might be funded by the Italian Antarctic agency, and suggested that the two groups – both of which had extensive experience with balloon-borne bolometric experiments – might collaborate on developing a telescope to fly over the Antarctic.

Though the attractions of the Antarctic flight were obvious – flight times measured in weeks rather than hours – the challenges were daunting, even in comparison with an ordinary balloon flight.  The payload would travel thousands of miles from the launch site and would thus need to operate fully autonomously. The telescope had to very reliably point towards the target region at all times to avoid catastrophic heating of the optics if exposed to the ever-present sun. All of the data would need to be stored on board and physically recovered at the end of the flight.  Computer systems had to operate reliably for weeks the elevated cosmic ray flux over the South Pole. The cryogenic system needed to be large enough to keep the detectors cold for weeks, yet sufficiently lightweight to fly on the balloon.

Lange and de Bernardis met in 1992 and sketched out a system that was well adapted to the challenging environment, but which was highly unconventional in one fundamental respect.  Relying on the breakthrough in detector stability that Lange had achieved in Berkeley, the new experiment would forgo the rapid modulation on the sky that had been a basic design principle in all past CMB experiments. Instead, each detector would stare at the sky as the entire telescope slowly scanned the target region.  This strategy reduced the cost, mass, and complexity of the experiment considerably and turned out to work extremely well.

BOOMERanG Takes Shape

It was not easy at first to obtain funding for BOOMERanG. The conventional wisdom said that the slow-scan approach would not work, and the competition for Antarctic flights was stiff.  Still, working with a small number of students, Lange and de Bernardis began to develop the basic technologies that would be needed.  The detectors and associated electronics were developed in the US, and the cryogenics, optics and attitude control system in Italy.  Other optical components came from the U.K. laboratory of Peter Ade, a world expert in the design and fabrication of mm-wave filters.  Soon after the development began, Lange moved to Caltech and began to develop detectors at JPL with his former student Jamie Bock.  Here things began to progress quickly, thanks to the support of Charles Elachi, then head of astrophysics and now director of the lab, and access to the world-class facilities of JPL’s micro-devices laboratory.

As funding was eventually secured from the Italian Space Agency (ASI), the Italian Antarctic program (PNRA), NASA and the US. National Science Foundation’s Office of Polar Programs, the team grew and the payload began to take shape.  After a final integration at Caltech, the 1000 kg telescope was shipped to Texas for a short test flight in summer 1997.

The details of the 1997 campaign convey much of the excitement and risk of ballooning.  After many weeks of integration in the hot, humid Texas climate, the first attempt to fly failed due to a leaky balloon that landed the telescope in a lake some miles from the launch site. After a good soaking, BOOMERanG was fished out of the lake and quickly refurbished. In a rather desperate attempt to fly again while the favorable winds persisted, the on-board computers were disassembled chip by chip, cleaned of mud and algae and reassembled.  Miraculously, one computer worked sufficiently well during the subsequent flight to allow the payload to function, though for a terrifying 30 minutes during the ascent it failed, only to come back to life when the team was about to abandon hope.

Hints of Flatness

A bit more than a year after the North America test flight, and while the data from the test flight was still being analyzed, the payload was equipped with a new focal plane and shipped to the Antarctic.  It launched on December 28, 1998, and landed 11 days later just 80 km away, after an 8000 km circumnavigation of the South Pole.  The performance was near-perfect – images of the galactic plane were shown at conferences while the payload was still flying, and cosmologists around the world watched the balloon’s progress as it journeyed around the Pole.

The data set was of an unprecedented size:  though only 4% of the sky was mapped, it was mapped with over 30 times the angular resolution of COBE, making for a map with almost 100 times as many pixels.  New analysis techniques were necessary in order to create an optimum map from the 10 days of data, and to properly tease out the power spectrum. 

The story would not be complete, at this point, without some mention of the intense competition between groups working towards the same goal of imaging the CMB using a variety of techniques. The geometry of the Universe, which inflation predicted to be exactly flat, could be measured by resolving the characteristic angular diameter of the structure in the CMB that COBE had detected.  The apparent angular diameter was a function of the physical size of the structures when the photons decoupled from matter and the geometry of the intervening space, which would either magnify or demagnify these structures if the space had positive or negative curvature, respectively. If inflation was correct and the geometry was flat, then the angular diameter would be approximately 1 degree. A number of experimental teams were racing to be the first to resolve the CMB and thus to both settle the decades-old debate about the geometry of the Universe and test inflation.

While the results of both the test flight and the Antarctic flight of BOOMERanG were still being analyzed, a group led by Lyman Page at Princeton and Mark Devlin at UPenn was analyzing data that had been gathered over several seasons using conventional radio receivers on Cerro Toco high (5200 m) in the Chilean Andes.  The “TOCO” experiment was the culmination of years of work with similar systems in Saskatoon and on balloon-borne platforms. In June of 1999, the TOCO collaboration published data showing a rise in power approaching an angular scale of 1 degree and a decrease at smaller angular scales.  The first hints of flatness had been seen.

Five months later, in November 1999, the BOOMERanG team published results gleaned from the test flight over 2 years earlier [Mauskopf et al. 2000]. Though this flight had been intended only to test the new technique, the power of the method was such that the few hours of data obtained in the test flight produced a result of comparable significance to the TOCO result.  In a companion paper [A. Melchiorri et al. 2000] the team published the first explicit declaration of a measurement of the geometry of the Universe, albeit still with modest (~ 25%) precision. 

First Resolved Images of the CMB

The first results from the Antarctic flight were published soon after, in April 2000 [de Bernardis et al. 2000]. While all previous detections of structure in the CMB, including the TOCO and North American BOOMERanG results had been statistical, the Antarctic flight of BOOMERanG produced, for the first time, high signal-to-noise images of the sky that showed the unmistakable signature of structure in the CMB with a characteristic angular scale of 1 degree, exactly as predicted by inflation. The detailed analysis of the data in the first release was limited to a small fraction of the entire data set. Even so, the angular power spectrum of fluctuations confirmed not only the flat geometry predicted by inflation, but also the adiabatic nature of the fluctuations, (via the width of the first peak in the spectrum), and the scale-invariance of the initial fluctuations (via the slope of the overall spectrum). [Lange et al. 2001] The data were consistent with inflation in every respect.

Within weeks of the first results from BOOMERanG, its sister experiment, MAXIMA – a successor to the MAX experiment developed at Berkeley, confirmed and extended the BOOMERanG results to even smaller angular scale [Hanany et al. 2000, Balbi et al. 2000].  MAXIMA, using detectors from Lange’s group and with an attitude control system from Italy, had used a conventional balloon flight over the U.S. desert to map ~ 1 % of the sky during a single night.  The agreement between the two experiments, viewing different parts of the sky, was convincing.  A decades old question had been decisively answered: the geometry of the Universe was flat. Inflation had passed its first test.

There was much more to be learned.  In April 2001, the first analysis of the full data set from the Antarctic flight revealed the harmonic overtones in the angular power spectrum that had long been predicted. [Netterfield et al. 2002, de Bernardis et al. 2002]. With more features in the power spectrum pinpointed, more of the cosmological parameters could be determined with precision.  The baryon density, previously estimated by comparing the measured abundances of light elements to the theory of big bang nucleosynthesis, was measured in this completely orthogonal manner to be in complete agreement, ending speculation that a good deal more baryonic matter could be tied up in dark cinders of an early generation of stars.  But this meant that the bulk of the matter in the Universe was in some exotic form never before detected. 

Taken together with data on the distribution of galaxies on large scales in the Universe today, the BOOMERanG images of the embryonic Universe revealed the basic recipe for the Universe. The images provided unambiguous and independent evidence that most of the matter in the universe was “dark” – of an exotic form never observed in the laboratory that did not interact electromagnetically at all – and, stranger still, that most of the energy density of the Universe was in the form of “dark energy”, a mysterious pressure that is causing the expansion of the Universe to accelerate. Though there was prior evidence for each of these strange ingredients (most notably rotational velocities of galaxies for dark matter and the apparent brightness of distant supernovae for dark energy), no single measurement had ever before revealed the entire recipe.

In capturing the first resolved images of the most distant objects that we will ever be able to see, the BOOMERanG results thus settled a decades old puzzle in cosmology.  Astronomers studying the relative abundance of light elements in the Universe had long concluded that normal matter could account for no more than 10% of the density necessary to make the geometry flat.  Astronomers studying the orbital trajectories of luminous matter had inferred that there was at least 25% of this critical density. Theorists had argued, however, that the density of everything must be exactly the critical density.  In the end all three were correct.  The Universe consists of ~ 5% ordinary matter, 25% dark matter and 70% dark energy. 

Simultaneous with the April 2001 results of BOOMERanG, the first results of the Degree Angular Scale Interferometer (DASI) were released. These also showed evidence for multiple peaks in the power spectrum, and constrained the basic cosmological parameters with a precision compable to BOOMERanG. The agreement between the two experiments was startling. The power spectra of temperature fluctuations recorded in different areas of the sky, at different frequencies (30 and 150 GHz), using completely different techniques were completely consistent with one another.

These results were to be confirmed several times over in rapid succession by other beautiful experiments that had been many years in planning and development:  In 2002, new results were reported from a yet more sophisticated analysis of BOOMERanG [Ruhl et al. 2003], as well as from the ARCHEOPS balloon-borne experiment [Benoit et al. 2003], the Cosmic Background Interferometer (CBI) in Chile, the Very Small Array (VSA) in Tenerife, and the ACBAR experiment at South Pole.  All were in good agreement, and all supported the same basic picture:  an Inflationary universe dominated by a mix of ~ 70% dark energy, 25% dark matter and 5% ordinary matter.  In early 2003, NASA’s Wilkinson Microwave Anistropy Probe (WMAP) released the results of its first year of survey of the entire sky, which improved the precision of the power spectrum at larger angular scales considerably.  The WMAP results further tightened the constraints on the various parameters, but without any significant change in the overall picture.     

The Future

Much remains to be done. All measurements of the faint but intricate patterns that are imprinted in the temperature distribution of the CMB are consistent with what is in some respects a simple model.  But this simple model contains 3 elements of which we must remain skeptical:  in increasing order of how skeptical we should be, these are Dark Matter, Inflation and Dark Energy.  Though there is compelling circumstantial evidence for each, none is well understood. It is possible that one or more will turn out to be in the same category as the “ether” that Michelson and Morley set out to detect a century ago; something that vanished in light of a new theoretical understanding. The need for these strange ideas is a hint that we may be about to learn something new and important. To make progress, we need to probe each idea more deeply.


The next big step in observations of the temperature fluctuations in the CMB is Planck, an orbital CMB telescope scheduled for launch in 2008 to an orbit 1.5 million km from the Earth. Planck has been developed by the European Space Agency with contributions of several key technologies from the U.S. In a real sense, Planck will be the culmination of the BOOMERanG, ARCHEOPS and MAXIMA balloon-borne experiments that we have together participated in.  Each of these experiments provided an essential demonstration of one or more of the technologies and methodologies that will be used on Planck. Planck will survey the entire sky with an unprecedented combination of angular resolution, sensitivity, and frequency coverage, producing the best map of the temperature anisotropy ever made.

By improving the precision with which we can measure most fo the cosmological parameters, Planck will carry tests of Inflation to the next level. Detailed models of inflation predict a subtle departure from scale invariance that Planck is well-designed to measure.  There are tantalizing hints of such a departure from scale invariance in the current CMB data [MacTavish et al. 2006].  Planck will provide a ten-fold improvement in the precision with which we can see this effect and, in doing so, will begin to differentiate one model of inflation from another.

Two new directions of CMB research promise to reveal much about Inflation and Dark Energy:  observations of the polarization of the CMB and observations of the secondary temperature anisotropies introduced on arcminute scales by astrophysical processes in the more recent history of the Universe.  Though Planck will provide important data on each, new ground-based and balloon-borne experiments now in development will provide important complements to the capabilities of Planck.


While the basic characteristics of the primary temperature anisotropies are now known, another property of the radiation – its polarization – has only begun to be explored. Like sunlight reflecting off of the surface of a lake, the CMB is polarized in a very specific manner by the physical processes that took place at the surface of last scattering.  This polarization provides a further property that can be used to more precisely constrain the basic cosmological parameters that are already probed by the temperature anisotropy.  Even more exciting, the special symmetries (“zero curl”) of the polarization allow it to be used as a tool to probe the detailed physics of inflation.

Exploiting the polarization of the CMB will be challenging.  The signals of interest are 10 to 1000 times fainter than the temperature anisotropies that took almost 40 years to resolve.  Spurred on by the enormous impact of the temperature measurements, however, the experimental community is making rapid progress in developing the technologies that will allow these extremely faint signals to be measured. Already, 4 experiments have detected polarization. DASI and CBI required only minor modification to detect polarization, and were the first to do so (in 2002 and 2003). 

BOOMERanG required a new type of detector:  a bolometer sensitive to mm-wave polarization.  The first such was invented at Caltech by William Jones and Lange in 2000 [Jones et al. 2006], and 8 were flown on BOOMERanG in 2003. Though the second flight suffered from a leaking balloon that caused the payload to sink slowly from 38 km to 20 km over the course of 10 days, the data was recovered and a detection of polarization of sensitivity comparable to DASI and CBI was published in July 2005 [Masi et al. 2006, Jones et al. 2006, Montroy et al. 2006, Piacentini et al. 2006].  WMAP followed with a detection on larger angular scales in early 2006.

While the polarization measurements have already begun to yield important cosmological information [cf. MacTavish et al., 2006], the most exciting application of CMB polarimetry is to directly probe the still unknown physics of Inflation.  Inflation occurs well behind the veil of opaque plasma that forever limits our view of the early Universe to an age of 380,000 years. To probe it we need to use something other than electromagnetic radiation.  The best candidate is gravitational radiation.  The enormous accelerations that occur during Inflation inevitably produce intense gravitational radiation to which the Universe is transparent.  Though this radiation has been stretched to wavelengths far greater than can be detected by terrestrial gravitation-wave detectors, recent theoretical advances show that it will leave a signature in the polarization of the CMB that may be detectable.  Put simply, the gravitational radiation produced during inflation will change the symmetry of the polarization in a unique manner.  The amplitude of the signature would directly measure the energy scale at which inflation occurred, revealing the physics responsible for the birth of our Universe.

We have some distance to go before we will be able to accomplish this.  In fact, the increase in sensitivity required is exactly that that separated COBE and BOOMERanG.  During the 1990’s we were able to achieve this by building more sensitive detectors.  This is no longer possible, as the detectors are already ideal.  To make progress now we need to build larger systems with thousands, rather than a handful of detectors.  One such detector system now under development at Caltech and JPL [Kuo et al. 2006] will be deployed to the South Pole [Yoon et al. 2006] and on high altitude balloons before the end of the decade. We hope very much to use half of the research portion of the Balzan funds to support young scientists in this research.

Secondary Anisotropy

The visible Universe is structured in a “comic web” of galaxies, organized in clusters and superclusters, and separated by giant “void” regions of lower density.  How did this structure form from the very homogeneous primeval plasma observed with the CMB ? High angular resolution observations of the Cosmic Microwave Background probe this process. For these studies, the CMB provides a homogenous background light, interacting with cosmic structures in different ways.  These interactions are called secondary anisotropies.

The largest secondary anisotropy is the inverse Compton scattering between CMB photons and free electrons in the hot intergalactic gas of galaxy clusters.  In fact, clusters shine in the X-rays because of the emission of this gas, which represents an important fraction of the mass of the cluster. There is a low (about 1%) probability that a CMB photon crossing the cluster interacts with an energetic electron. The electron will pass to the photon a small fraction of its energy: the outgoing photon will have a higher (about 1%) frequency. This process removes photons from the low-frequency side of the CMB spectrum, and adds photons to the high-frequency side. The net result of the process is a very characteristic distortion of the spectrum of the CMB in the direction of clusters of galaxies, known as the Sunyaev-Zeldovich effect. Being a scattering effect, its amplitude (of the order of 10-4 – 10-5 of the CMB brightness) does not depend on the distance of the cluster. So, in principle, it is possible to detect extremely distant clusters of galaxies, mapping the large-scale structure of the universe and understanding when clusters formed and how abundant they were in the past. This, in turn, is related to the expansion rate of the Universe and to the presence of Dark-Energy. Moreover, combining SZ observations with observations of the X-ray brightness it is possible to estimate the linear size of the clusters, thus building a new kind of Hubble diagram to measure the expansion rate of the Universe (the Hubble constant).

The Sunyaev-Zeldovich effect has been observed in several clusters, using radio telescopes, as a reduction of CMB brightness at low frequencies. Due to the effect of atmospheric emission, the observation of the increase in CMB brightness at high frequencies is more difficult, and has been done only for a few clusters.

For a firm detection of the effect and a clean separation from competing foregrounds, the experiment should detect simultaneously both the decrease and the increase of CMB brightness. The Rome group has built a large balloon-borne telescope devoted to this observation [Masi et al. 2005]. The OLIMPO instrument is able to map, with arcminute resolution, selected patches of the sky, at wavelengths of 2 mm (negative SZ effect), 1.4 mm (null effect), 0.9 mm and 0.5 mm (positive SZ effect). The experiment will carry out deep integrations in the directions of tens of clusters in a single long duration balloon flight. OLIMPO will also carry out a blind survey of a blank region of the sky. This will allow to search for distant clusters, undetectable in X-rays. Also, any other source of frequency-dependent anisotropy, like the resonant scattering of CMB photons against heavy elements in the cosmic structures, will produce interesting signals in the OLIMPO maps. The instrument is nearly completed, with the first flight scheduled for 2008. We hope very much to use half of the research portion of the Balzan funds to support young scientists in this research.


High resolution images of the CMB first captured with BOOMERanG and now many other experiments have revolutionized cosmology. The geometry of the Universe has been measured to be approximately flat.  The angular spectrum of temperature fluctuations leads us to believe that observable Universe underwent a superluminal “inflation” from sub-atomic to macroscopic dimensions, and that ordinary matter accounts for only a small fraction of the total mass energy density of the Universe. As many new questions have been raised as have been answered.  What it the nature of the dark matter and dark energy that dominate the Universe?  What physics drove inflation ? How did the cosmic structure develop from the unstructured early universe ? New technologies will soon allow us to use the CMB in new ways to address these questions – the story is not over yet.


BOOMERanG, MAXIMA and ARCHEOPS each involve dozens of extremely talented scientists and engineers who jointly share the credit for the their success. We are extremely thankful to have had the opportunity to work with these remarkable teams.  The work that we have described would have been impossible without the generous support of many government agencies. Major support in the U.S. came from NASA, from the Columbia Scientific Balloon Facility, from the Jet Propulsion Laboratory, from the NSF Office of Polar Programs, and from NERSC.  In Italy, major support came from PNRA (Programma Nazionale di Ricerche in Antartide), ASI (Agenzia Spaziale Italiana), University of Rome La Sapienza.


P. Meinhold et al., “Measurements of the Anisotropy of the Cosmic Background Radiation at 0.5o Scale Near the Star Mu Pegasi”, The Astrophysical Journal, 409, L1, (1993).

J. Gundersen et al., “A Degree Scale Anisotropy Measurement of the Cosmic Microwave Background Near Gamma Ursa Minoris”, The Astrophysical Journal 413, L1, (1993).

M. Devlin et al., “Preliminary Results from the Third Flight of the Millimeter Anisotropy Experiment (MAX)”, Proceedings of the National Academy of Sciences, 90, 4774 (1993).

A. Clapp et al. “Measurements of Anistropy in the Cosmic Microwave Background Radiation at Degree Angular Scales Near the Stars Sigma Hercules and Iota Draconis”, The Astrophysical Journal, 433, L57 (1994).

 S. Tanaka et al. “Measurements of Anisotropy in the Cosmic Microwave Background Radiation at 0.5° Scales near the Stars HR5127 and Phi Herculis”, The Astrophysical Journal, 468, L81 (1996).

G. Dall’ Oglio, P. de Bernardis “Observations of Cosmic Background Radiation Anisotropy from Antarctica” The Astrophysical Journal, 331, 547, (1988).

P. de Bernardis, et al.,  “A short wavelength measurement of the Cosmic Background Radiation Anisotropy” The Astrophysical Journal Letters,  360, L31, (1990)

P. de Bernardis, et al. “Degree-scale observations of cosmic microwave background anisotropies” The Astrophysical Journal, 422, L33, (1994).

S. Masi, et al. “Foreground removal and CMB fluctuations in a multiband anisotropy experiment (ARGO 1993)”, The Astrophysical Journal, 463, L47, (1996).

P. D. Mauskopf, et al. “Measurement of a Peak in the Cosmic Microwave Background Power Spectrum from the North American test flight of BOOMERANG” The Astrophysical Journal, 536, L59 (2000).

A. Melchiorri, et al. “A measurement of Omega from the North American test flight of BOOMERANG” The Astrophysical Journal, 536, L63 (2000).

P. de Bernardis, et al. “A Flat Universe from High-Resolution Maps of the Cosmic Microwave Background Radiation”, Nature, 404, 955 (2000).

A. E. Lange, et al. “First Estimations of Cosmological Parameters From BOOMERANG”,  Phys.Rev. D 63, 042001 (2001). 

S. Hanany, et al. “MAXIMA-1: A Measurement of the Cosmic Microwave Background Anisotropy on angular scales of 10 arcminutes to 5 degrees”, The Astrophysical Journal, 545, L5, (2000).

A. Balbi, et al., ”Constraints on Cosmological Parameters from MAXIMA-1” The Astrophysical Journal, 545, L1, (2000).

C.B. Netterfield, et al., “A measurement by BOOMERANG of multiple peaks in the angular power spectrum of the cosmic microwave background”, The Astrophysical Journal, 571, 604, (2002)

P. de Bernardis, et al., “Multiple Peaks in the Angular Power Spectrum of the Cosmic Microwave Background:  Significance and Consequences for Cosmology” The Astrophysical Journal, 564, 559, (2002).

J.E. Ruhl, et al., “Improved Measurement of the Angular Power Spectrum of Temperature Anisotropy in the CMB from Two New Analyses of  BOOMERanG Observations”, The Astrophysical Journal, 599, 786 (2003).

Benoit A., et al.,  “The Cosmic Microwave Anisotropy Power Spectrum measured by Archeops”, Astronomy and Astrophysics, 399, L19, (2003).

S. Masi et al. “Instrument, Method, Brightness and Polarization Maps from the 2003 flight of BOOMERanG” , Astronomy and Astrophysics,  458 , 687-716, (2006)

W. Jones et al., “A Measurement of the Angular Power Spectrum of the CMB Temperature Anisotropy from the 2003 Flight of Boomerang” The Astrophysical Journal, 647, 823, (2006).

C. MacTavish et al., “Cosmological Parameters from the 2003 flight of BOOMERANG” The Astrophysical Journal, 647, 799, (2006).

F. Piacentini et al., “A Measurement of the Polarization-temperature Angular Cross Power Spectrum of the Cosmic Microwave Background from the 2003 flight of BOOMERANG” The Astrophysical Journal, 647, 833, (2006).

T.Montroy et al., “A Measurement of the CMB <EE> Spectrum from the 2003 Flight of BOOMERANG”, The Astrophysical Journal, 647, 813, (2006).

W. Jones et al., “Instrumental and Analytic Methods for Bolometric Polarimetry”,

C.L. Kuo et al. “Antenna-Coupled TES Bolometers for CMB Polarimetry”,

K.W. Yoon et al., “The Robinson Gravitational Wave Background Telescope (BICEP): a bolometric large angular scale CMB polarimeter”,

S. Masi et al., “A balloon-borne survey of the mm/sub-mm sky: OLIMPO”, Proc. Of the 17th ESA Symposium on European Rocket and Balloon Programmes, ESA-SP 590, 581-586, (2005).

Insert your email address here to sign up for our newsletter with all the latest updates on the International Balzan Prize Foundation .

By subscribing, I confirm that I have read and understand the General Terms and Conditions of EU Regulation 2016/679, in particular with regard to my rights, and hereby consent to the processing of my personal in the manner and for the purposes indicated therein.
Fondazione Internazionale Premio Balzan