tisdag 9 december 2014

How often does one have to kill the Cat?

Will a new report help to finally close the strange case of the so-called Energy Catalyzer, the E-Cat, of Andrea Rossi?
The E-cat has been around for almost four years now. Claims have been made but were never proven. Promises have been made but were never fulfilled. A report by Levi et al. has been written and heavily criticized in 2013. And instead of answering [1, see also edit below] to the critique and questions raised, new measurements with yet another version of "the" E-cat have been made and a new report, the so called Lugano report by Levi et al., was spread 2014 on the internet [2]. What was "indications" in 2013 was now claimed with certainty [3]. Still, of course, some call for further studies [4].

Use Google and you will find several places on the internet where critical reviews of the current (or any of the previous) claims in the E-Cat story have been published. Here, I just want to mention the recent critique by Christian Schumacher (Part 1 and Part 2) and the excellent comments by Ethan Siegel here and here [5]. 

On this blog I have raised the issue of isotope abundance claimed by Levi et al. [6]. I have commented on Rossis extraordinary capability to, despite his changing and contradictory claims, keep his followers happy. And I have written a bit more broadly about the role of science and critical thinking in relation to this case.

For me as a nuclear physicist the claimed very drastic change of the isotopic abundances alone is a smoking gun. And the the cat is dead. For people more focusing on the electrical side and wondering how things might really be connected the new, very well-written and well-argued report by Per Rutquist is even hotter. The kind of analysis Rutquist delivers is what is called for when discussing the E-Cat, but which is - surprisingly - lacking in the work by Levi et al.

I bet believers will continue to trust in Rossi and continue to find excuses [7] but the many critical points found and raised show that Levi et al. - although they must have been well aware of both the public attendance and scrutinization that their report will receive - have shown rather poor craftsmanship and, again, delivered a report that has too many blind-spots to be taken seriously.

So, well, maybe the Cat is actually a vampire. Certainly sunlight into some still closed boxes would help. But even without sunlight, the report by Rutquist is the wooden stake that really finishes the Cat off.

[Edit 2014-12-09 20:55: And yes, concerning not giving answers: It is noteworthy that two of the authors of the Lugano report, Hanno Essén and Bo Höistad, have been contacted by Rutquist five weeks ago, i.e., November 3. But, he has still not received any answer.]


[1] I do not count this interview on ecatnews.com as an answer were Bo Höistad expresses his opinion about the critiques ...
[2] Financed by e.g. Elforsk. Elforsk also published a strange and pretty low-level report on the LENR here[3] Just read the title "Observations of abundant heat production from a reactor device and of isotopic changes in the fuel".
[4] See statement by Elforsk here and a critical comment on Elforsks uncritical attitude here.
[5] A brief critical piece has recently been published on Spiegel online (in German).
[6] Or, as Rossi in true marketing spirit likes to refer to the authors: "the professors".
[7] This is all to natural. Read about the mechanism here.

torsdag 23 oktober 2014

The Pomp factor in Cold Fusion - a reply to the open letter by Patrik Wiksten

I have been honored with an open letter by Patrik Wiksten titled “The Pomp factor in Cold Fusion, an open letter to Stephan Pomp” on lenr-forum.com. Of course this well formulated letter deserves an answer so I’ll give it a try. 

The fact that Wiksten has dedicated his reflections about the question of the E-Cat and the continued testing to me probably means that I have touched on a sensitive issue for cold fusion [1] and the understanding of how science progresses in my recent blog-post “The Cat is dead” [2].

Let me start with stating that I think I have an obligation to take part in the E-Cat discussion. I am employed by Uppsala University – and thereby in a sense by the Swedish tax-payers – as professor in applied nuclear physics and as such I teach nuclear physics on various levels. In addition, I also give lectures on the topic of “Science and Pseudoscience” in schools, to students, and on the PhD student level. 

Given the additional fact that the E-Cat-story, since it began to come to public attention in January 2011, involves people from Uppsala University, it is clear that it would be wrong for me not to honor a university teacher’s duty within the so-called 3rd task. This 3rd task means that I should not only teach and do research within the walls of the university, but also actively spread scientific knowledge to society. The latter very much includes to take part in discussions about science and scientific reasoning in general and to defend science.   

All this does, of course, not mean that I am failure free [3]. Far from that! And I think that I am very much aware of this quite normal human trait [4]. Actually, the ability to be self-critical and discard ones opinion/belief/previously acquired knowledge when confronted with new evidence is, arguably, at the very heart of what it means to be a scientist. This is a very hard thing to do and it is all too human to defend ones position despite facing contradictory facts.   

Now, how does our work as experimental nuclear physicists look like? When experiments are designed and run, data are analyzed and interpreted, many mistakes are made. This comes almost as a necessity. What we scientists do is supposed to be difficult since we try to extend the realm of knowledge and find out things that no one has reported before. Hopefully though, my colleagues and I manage to get rid of these mistakes by the time we publish the results. We look over and over what we have done, discuss if we think everything was done correctly, whether there might be any mistakes left, and whether there might be alternative explanations for our findings than the ones we propose. What we finally publish represents as complete information as possible for anyone to redo the experiment and/or at least check our analysis routines and judge if the presented conclusions can be drawn from the experimental findings.

A lot of the analysis work is devoted to this tedious and sometimes annoying thing called uncertainty estimation or error analysis. All to allow to challenge the reported results and possibly falsify them.

Because this is what we scientist generally strive for: to either improve older results or falsify them, providing new input as a challenge for theoretical physicists to explain our experimental findings. How boring it would be to just replicate what is already known and has already been done! I mean: where would the fun in science be if we could not challenge the results of the previous generations? Progress is made by identifying mistakes.

Now, Patrik, the mistake you seem to make is to believe that everything that challenges our current knowledge is correct just because it goes beyond “the current paradigm”. Well, we certainly wish for new knowledge but we have to test claims against the facts and things we know. There is a fitting saying: “You have to be open but not so open that your brain falls out!”

It seems obvious to me that the mistake of Levi et al. and, likely, of many people working on “LENR” is the confirmation and publication bias; when some measurement seems to confirm your believe (“excess heat”) it is assumed to be true and factual, and it gets published (or, rather, spread as news on the internet).

When, on the other hand, an experiment fails to produce the result one is looking for (i.e., everything looks as if it fits the current “paradigm”), nothing is published and one looks for a mistake in the setup.

You see the problem, Patrik? An observed excess heat (within LENR research) is an extraordinary claim and one should put in a lot of effort to check and double-check if everything was done correctly, can be reproduced, etc. In the E-Cat case, e.g., one seems, among other things, to have forgotten to check some cabling …

So what is the point of this long post? It simply is that what Levi et al. do is not well-performed research. One can even ask if it is justified to label these reports as “research” (see especially the Appendix, p. 11ff, here).

In addition, the findings reported by Levi et al. - surprisingly and most embarrassingly supported by Elforsk - do not call for an extension of nuclear physics. They outright contradict what we have learned the past 100+ years; no radioactive products, no radiation in the process, yet complete conversion of all nickel isotopes into just one (Ni-62 content being raised from 3.6% to about 99%). This cannot (!) be explained by some new reaction alone if you not also claim that previous fundamental knowledge is plain wrong!

So maybe I am just being lazy but I prefer to use Occam's razor and assume that someone has played a simple trick and switched the samples (already before or after the test). And while E-Cat-fans try to figure out how to reproduce the results without using Joe Labero I grab a beer, try to explain in blog posts how real science works, or just watch a good movie.


[1] That is if you claim (and some do) that E-Cat = LENR. If that is true (not impossible) than I am afraid to say that LENR is a dead Cat.
[2] See also the follow-up “Mr. Rossi, I admire you
[3] At least according to my wife and kids. But, of course, they might be wrong :-)
[4] My presentations about “Science and pseudoscience” contain quite a bit of material on this issue and psychological aspects on why we sometimes believe strange things.

torsdag 16 oktober 2014

Mr. Rossi, I admire you!

Yes, I do! You have accomplished something quite remarkable. For many years now you have presented various versions of "the" E-Cat and argued along different and contradictory lines. But none of this bothers your followers. Somehow you have this rare ability to make people forgive you even if you outright admit that you have cheated the people that you worked together with. Here is a screenshot and the cut-and-paste quote from your recent statements published under the heading "Rossi Responds to Swedish Professors Critical of E-Cat Report" on e-catworld.com and in your own "Journal of Nuclear Physics" (sic!):


The remarkable thing is that there is no outcry of indignation among the E-Cat believers. People forgive you and keep believing in you and "the" E-Cat. They even keep inventing more or less far-fetched excuses on your behalf. This must be due to the fact that the E-Cat has become a religion![1]

For let's be precise: you claim that Sven Kullander[2] had been "warned" about the manipulated sample. If that were true, why would anyone do any kind of analysis on the "fuel/ash"? Where is any statement of Sven Kullander or Hanno Essén or Mats Lewan that they were aware of this? No scientist would accept these conditions (i.e., analyzing a manipulated sample) without clearly stating that the sample is not "the real thing"! I sure would like to hear from Hanno Essén, Bo Höistad, and/or Mats Lewan if they actually knew about the manipulation and for some reason kept this important piece of information for themselves or if your statement has taken them by surprise!

Mats Lewan writes in his book with the very true title "An impossible invention":
"During his Uppsala sojourn Rossi left two small bottles of the fuel powder used in the E-Cat—one with unused fuel and the other with powder that Rossi said had run in the device for months. Later Kullander had measurements made on the powder, indicating that Rossi’s theory was wrong."
Mats Lewan. An Impossible Invention (Kindle Locations 1707-1709).
Nowhere can I find any mention that Sven was aware of the "incompleteness" of the sample. And of course we can no longer ask him to confirm or deny. What we have, though, is some emails from him and the previous report from Naturhistoriska Riksmuseet. The report makes no mention of any "warning" that the sample is incomplete. Sven wrote to Göran Ericsson on December 22, 2012:
Hej Göran

Bifogar rapporten från Naturhistoriska som gjordes under våren.
I den framgår klart att nickel inte är med i någon kärnreaktion.

Hsn Sven
[Translation: Hi Göran // Enclose the report from Naturhistoriska which was done this spring. // It clearly says that nickel is not involved in any nuclear reaction. // Grtngs Sven]
If Kullander had known about your claimed "warning", he would have understood the importance of conveying this information to his collaborators. Certainly he would also have pointed this out in his email and it would have been mentioned in both the report and elsewhere on the internet where the results were discussed at length.
Also the people from Ralon and KTH that studied the manipulated sample - and clearly showed that the copper could not have been produced from nickel in a nuclear transmutation process in the E-Cat (of that time) - make no mention of any manipulation. Instead, they, e.g., write:
"Prover som erhölls från Sven Kullander i december har analyserats. Proverna bestod av två flaskor med ca 1 gram i varje flaska. Ena flaska kallas ”NY” och innehöll det nickel pulver som Rossi använde i reaktorn, taget innan någon aktivitet har skett. Den andra flaskan kallas ”GAMMAL” och det innehöll pulver som använts i en av Rossis reaktorer i ca 6 månader. Denna flaska innehöll också ca 1 gram pulver."
[My translation: "Samples obtained from Sven Kullander in December have been analyzed. The samples consisted of two bottles containing about 1 gram each. One bottle is labeled “NEW” and contained the nickel powder that Rossi used in the reactor, obtained before usage. The other bottle is labeled “OLD” and it contained the powder that was used in one of Rossis reactors for about 6 months. This bottle also contained about 1 gram of powder."]
Curt Edström and Jan-Erik Nowacki, "Analys av två typer av nickelpulver", Ralon and KTH, 2013-01-17.
This is all. No mention is made that they, or Sven Kullander, knew that the analyzed and discussed sample has been manipulated.

Anyway: if what you state is true, i.e., that you removed (!) something from the sample but it actually was in the "reactor" I am even more baffled. It would mean that we had a different kind of nuclear reaction than we have now? No change in the isotopic composition then but a lot of change now? I mean, you cannot get back to natural isotopic composition by removing (!) something from the sample?

All this leaves only one conclusion: you were playing tricks then (trying to give the impression that copper was produced) and you are playing tricks now (trying to have people believe all nickel somehow converted into Ni-62). 

Your statement about the 2011 sample simply is an after-the-fact construction desperately trying to save you from the mistake of going from one extreme (no isotopic change in Ni and Cu but claims of nuclear reaction) to another, contradictory extreme (complete isotopic change in nickel to consist of only Ni-62 without any (!) other of the well-known nuclear reactions and without inducing any radioactivity).[3,4,5]

And the funny and very interesting thing is that this desperate try to justify and explain away all contradictions seems to work with your followers! They accept your claim that you need to do all this to protect your "patent". And you simply make good use of the fact that people want to believe in miracles. So you get away with it. Again. Well played! This is pure genius and I admire you!

Warm regards,

  Stephan Pomp

P.S.: Some earlier posts on the subject concerning Swedish media and their treatment of the subject are found here (SVT) and here (P1) and a comment related to the P1 reports is given here (all in Swedish).

Regarding the copper in the 2011 sample, Lewan writes:
"What was particularly controversial was the scientific content—a theory that the device produced energy via the fusion of hydrogen and nickel nuclei. The result of such a nuclear reaction would be copper, of which Focardi and Rossi had found traces in the fuel powder after use."
Mats Lewan. An Impossible Invention (Kindle Locations 1303-1305).
"As in the Fleischmann and Pons experiment, in Rossi’s device initial phenomena brought to mind fusion, including the discovery of copper in the nickel powder used as fuel. In this case, it could be the nucleus of nickel that had reacted with the nucleus of hydrogen, consisting of a single proton, which then formed a new nucleus, copper, because copper has one proton more than nickel—a fusion reaction that in itself would release energy if it occurred. But analysis of the used nickel shows that the copper was simply a contamination powder from another source."
Mats Lewan. An Impossible Invention (Kindle Locations 5113-5116). 
[1] Maybe this explains why the Cat has so many lives?
[2] As I mentioned in this blog post (in Swedish) - reporting from a talk Sven Kullander gave on Novermber 9, 2011 - Sven seemed to have no doubt that the "fuel/ash" was the real thing. And I remember having been very critical about this blind trust and believe (as the blog post documents).
[3] It should be pointed out here that I have full confidence in the different analyses of the isotopic compositions etc of the "fuel" and the "ash" that have been performed by various labs. These analyses were really independent!
[4] I have a suggestion for further improvement of the reported results in the next round (yes, I am pretty sure there will be a next round with a new report calling for more tests and money from, e.g., Elforsk ...): change the isotopic composition a little less dramatic and make sure that there is some residual radioactivity in the "ash".[4] Then it will be harder for us critics to claim that we know your game.
[5] Both you and many others have pointed out that it was the team of researchers that have taken out the ash. Maybe so and maybe all researchers are really doing their best to find out about the mystery of "the" E-Cat. But who put the sample in? Was there only one sample inside? Or was the sample that was removed by the researchers only one that the researchers believed was the same that first showed natural isotopic composition in Li and Ni?

onsdag 8 oktober 2014

The Cat is dead

You all know about Schrödinger’s cat. It is dead as soon as you take a look at it. A similar thing can finally be said about the E-Cat after having read the essential parts in the new report by Levi et al1. 

The title of the report, though, heralds quite differently “Observation of abundant heat production from a reactor device and of isotopic changes in the fuel”. So what is going on? 

Yet another version of the E-Cat2 has been tested. This time the tests have been performed in Lugano3.

As in previous reports, some measurements and technical details are reported in great detail. One may, however, wonder why so much information, irrelevant for the core question, is reported while the really interesting claim is dealt with on only about two pages.  

So if you do not manage to read through all the tables and numbers, just turn to page 27 ff, read section 8 “Fuel analysis”, and check out table 1 in Appendix 3. That should do. Why? Because none of the measurements presented on the previous 26 pages matter, if what is written in this section is true; i.e., that the reported dramatic changes in the isotopic composition of the “fuel” are really due to a nuclear reaction in the E-Cat.

Levi et al. write that the “fuel” initially consists of a mixture of nickel powder and lithium in natural isotopic compositions. However, after the run, the “ash” is radically different in the isotopic composition! Practically all Li-7 has turned into Li-6 and all the 4 other naturally occurring nickel isotopes have practically vanished and turned into Ni-62. The latter has a natural abundance of 3.6 % but in the “ash” the abundance is about 99 %! Yes, you have read correctly. This is what is claimed. Nobelprize? If true: definitely. Imagine: You run the E-Cat and all the Ni-58 (68 % natural abundance), Ni-60 (26 %), Ni-61 (1 %) and Ni-64 (1 %) nuclei have turned into Ni-62.

Yes, you may read this again and try to digest it. The authors really claim that some of the nickel isotopes get some neutrons added while others have some removed and everything just becomes one single isotope. 

And this miracle happens without any radiation being emitted when the E-Cat is run, without traces of copper or other elements, and without changes in the effectiveness of the E-Cat while it is run4.

Levi et al. draw the conclusion that “nuclear reactions have taken place” and that one “can speculate about the nature of such reactions.” However, they “refrain from such discussions.” While the latter seems wise it is totally inexplicable to me, how the authors cannot see the most obvious and by far most likely conclusion of the fuel analysis; that they simply have been fooled. Just realize that obtaining an enriched Li-6 or Ni-62 sample is not too difficult (see, e.g., here, here and here). And yes, the available enrichments in Li-6 do match what is reported …
So for my part the Cat is dead. How others accomplish not to see and obviously are able to keep the Cat alive in their wishful thinking, that is the real mystery.

1) The current report is said to have been submitted to arXiv again but has been put on hold there. In addition it seems to even have been submitted to a real physics journal. I would be very surprised if it got published. When taking a look at the quoted references what sticks out to me is the lack of a reference to the critique of the earlier report by Levi et al. However, references to, e.g., Wikipedia are made. This is quite unusual for a paper submitted to a scientific journal.
2) I have lost count on how many different versions there are as it seems never to be the same that is tested twice.
3) It is noteworthy that TSL is no longer involved and several authors of the previous report are not among the authors any longer. The only connection to Uppsala University is now through three retired researchers. The extensively reported measurements of the radiated power have been carried out in southern Switzerland. 
4) Think about it: the E-Cat is run for what seems to be an arbitrary amount of time under what is reported as stable condition and in the end all the “fuel” has turned into “ash”.

fredag 30 maj 2014

Rossi-affären i P1 - en kommentar

Under sista veckan i maj körde P1 flera program i Vetandets Värld som granskar Rossi-affären och de svenska journalisternas och forskarnas inblandning (Medierna, Vetandets Värld del 1, del 2, del 3) och jag skrev kort om detta här.

Med anledning av denna granskning fick jag först en förfrågan om att vara med om en intervju med Vetenskapsradion men man beslöt sedan att hellre prata med en ”neutral” person, alltså någon som inte redan tydligt positionerat sig (dels i P1 programmen och dessutom t ex här, sen här, och senast här och här). Synd men helt förståeligt. Frågorna som var tänkta att diskuteras tar jag därför upp här och försöker svara på de ur min synvinkel [edit: den utmärkta kommentaren från Vetenskapsradios chef och seriens producent Ulrika Björkstén finns här].

Hur kan sådant här hända? Hur lurar en tidigare dömd bedragare bland annat framstående svenska forskare att stödja något som med all sannolikhet är fusk?
Det som driver en forskare är både nyfikenhet och hoppet på upptäckter. Att känna att man kanske är med på någon stor upptäckt är en fantastisk känsla och säkert en av drivkrafterna bakom. 

Vad som tillkommer är att några av de inblandade forskarna redan var aktiva inom området (Sven Kullander [1]) eller har tidigare skapat teorier som skulle kunna förklara kall fusion (Hanno Essén [2]).

Sen, när man väl har gett sig in i E-Cat undersökningarna och börjat ge Andrea Rossi (och Guiseppe Levi!) sitt förtroende kan det vara svår att ta sig ut igen. Trots möjliga tvivel och trots misslyckanden i undersökningarna. Eller kanske till och med på grund av misslyckanden. Detta kallas cognitive dissonance. Man börjar hitta på ursäkter eller bortförklaringar för sig själv och mot omvärlden som leder till djupare tro på Rossi och hans E-Cat och minskat kritiskt tänkande.  

Men den kanske allra viktigaste faktor verkar vara personligheten Andrea Rossi. Själv har jag aldrig träffat honom, det medger jag gärna, och kan således inte bedöma just den aspekten. Men det verkar troligt att hans sätt att handskas med folk, och då speciellt beundrare, gör att man känner stort förtroende för allt han säger hur inkonsistent det än må vara och hur ofta han än ändrar sina beskrivningar och påståenden. Allt går att bortförklara om man vill.

En del av mekanismerna beskriver Michael Shermer i sin bok Why people believe weird things där det i den senaste upplagan finns ett extra kapitel med titeln Why smart people believe weird things. En central del i Shermers förklaring är att välutbildat folk har bra träning i att försvara sina positioner. Detta kan göra det lättare att hitta på ursäkter som låter trovärdiga, till och med för de själva.

Var går gränserna för vad man kan hålla på med på ett universitet? Kan fri forskning vara hur galen som helst?

Svaret på den senare frågan är "Ja!". Bara det är fackmannamässigt genomfört. Idéer kan verka vara helt galna från början men sen visa sig vara bärkraftiga och möjligen vara viktiga forskningsgenombrott. Men som sagt, forskningen måste vara fackmannamässigt genomfört, vara under forskarens kontroll osv. I Rossi-affären är det inte givet från början att saken är helt galen och aldrig kan fungera. Problemet är att Rossi hela tiden har full kontroll. Han styr hur mycket de inblandade får veta, ändrar förutsättningarna för planerade tester osv. Detta gör att det inte handlar om forskning längre men att forskarna som vågade ge sig in i denna röra blir till verktyg för Rossi och hans marknadsföring. När (och om!) de inblandade inser detta kan det vara för sent och man tar till olika försvarsmekanismer för att rättfärdiga det man har gjort. Det är en svår konst att medge att man ha misstagit sig och att man har investerat mycket tid för att hjälpa en bedragare.

Vad gäller den första frågan så är det så att det finns gränser så klart. Främst kanske inom medicin och biologi. Det krävs etiska prövningar om forskningen involverar människor eller djur. Men även inom andra forskningsområden kan det finnas etiska själv för att inte forska på vissa områden och det kan finnas begränsningar av säkerhetsskäl (exempelvis hantering av farliga ämnen).

Men som sagt, annars finns det inga begräsningar för att använda sin fria forskningstid på i princip vad som helst. Och det är så det ska vara på ett universitet.

Vilka kontrollmekanismer finns/borde finnas för att skydda sig mot bedragare?

Sunt förnuft och en del skepsis. Om det låter för bra för att vara sant så är det oftast det, lyder ett ordspråk. Carl Sagan pratar om ett skeptisk och vetenskapligt förhållningssätt som han kallar ”baloney detection kit”. Vad detta är beskriver Michael Shermer utmärkt i en 15 minuter lång video. Det handlar om tio frågar som man kan ställa sig för att bedöma om det är någonting seriöst man har att göra med eller om det snarare är ”baloney”. Viktiga sådana testfrågor är

  • Hur tillförlitlig är uppgiftslämnaren? 
  • Har påståenden verifierats av andra?
  • Har någon försökt motbevisa påståenden? 
  • Finns det andra, enklare förklaringar? 
  • Hur öppet redovisas viktig information? 
  • Är det rimligt med tanke på hur världen fungerar?

Det senare kan tillämpas på exempelvis Nigeriabrev och löften om fri och ren energi från E-Cat där man endast behöver ha en hemlig ingrediens.

Att lära sig ett vetenskapligt förhållningssätt är en central del av forskarutbildningen, alltså det kritiska tänkandet. Framförallt lär man sig (eller bör man lära sig) att använda det kritiska förhållningssättet även - och kanske främst - på de egna resultaten. Detta är det kanske svårarste man måste lära sig som forskare: Att bli glad över upptäckta fel. För det är ju så man kommer vidare i forskningen. Man rättar till fel, förbättrar, och går vidare. Det är ju sällan eller aldrig man gör allt rätt från början speciellt när man ska forska fram helt nya resultat, där det alltså inte finns redan färdiga svar! Att fela är mänskligt men att medge att man haft fel kan vara hur svårt som helst. Det har vi nog alla en erfarenhet av.

[1] Sven Kullander har under många år stött Hidetsugu Ikegami och hans forskning kring vad de kallar chemonuclear fusion, se t ex sid 3 i denna dokument och sista stycket i en text av Hanno Essén här. Denna forskning har genomförts med finansiellt stöd av bl a Energimyndigheten (se här och här). Projekthandläggare för P20628-1 (se rapport EFA-05/2, "Observation of enormously enhanced nuclear fusion in metallic Li liquid" och EFA-05/3) vid Energimyndigheten var för övrigt Lars Tegnér (en av medförfattarna av den omdiskuterade arXiv rapporten) som sedermera blev hedersdoktor (2008) vid Uppsala Universitet. 

[2] Se t ex här.