Relying on new affidavits from leading experts on thermonuclear weaponry, the Justice Department yesterday renewed its warning that publication of a proposed magazine on the workings of the hydrogen bomb "would irreparably impair the national security."

The government also rejected contentions by the progressive, a monthly, that the Atomic Energy Act of 1954 is unconstitutionally broad because it applies to data that were the magazine's original work product and not acquired from any government source.

"Congress properly determined that, in order to protect national security, the confidentiality of data pertaining to atomic weapons must be protected regardless of the source of the data, at least where [someone] would have reason to know that disclosure would harm the United States or give an advantage to other nations," the department said.

The department made its arguments in a statement filed in U.S. District Court in Milwaukee, where it is seeking a preliminary injunction to prevent the Progresive from publishing -- possibly in its May issue-- the Hbomb article. The writer is Howard Morland, 36, a free-lancer who spent six months researching the article.

On Friday, Judge Robert W. Warren, acting in a case with apparent major implications for freedom of the press as well as for national security, granted a government petition for a temporary order to restrain publication.

Tomorrow the Progressive will file its answers to the new government arguments, probably along with affidavits from scientists who, like the department's experts, also have read the article but who disagree that it could imperil national security.

On Friday, Warren will hear oral argument in the case, which ulitmately may reach the Lupreme Court.

One of the new affidavits came from nuclear physicist Jack W. Rosengren, who was responsible for the physics design of the first Polaris warhead and an early Minuteman warhead. Later, he became associate director for nuclear weapons design at the Lawrence Livermore Laboratories in California, and then deputy director for science and technology at the Defense Nuclear Agency.

"The Morland article goes far beyond any other publication in identifying the nature of the particular design used in the thermonuclear weapons in the U.S. stockpile," Rosengren said.

The article "describes in a relatovely detailed manner the basic design concepts and certain specific design features of U.S. thermonuclear weapons," he said.

"My appraisal... is that it contains a significant amount of information which is properly classified as Secret Restricted Data. While certain bits and pieces... have been previously disclosed in scattered sources, the article nevertheless contains information that has never been released anywhere before.

"It provides a more comprehensive, accurate and detailed summary of the overall construction and operation of a thermonuclear weapon than any publication to date in the public literature.

"There are many feasible and grossly different possible designs" for H-bombs, some of which are "awkward," require "impractical amounts" of scarce materials, or cannot be made to operate efficiently, Rosengren said.

Nowhere in the open literature, however, is there "a correct description of the type of design used in U.S. weapons," Rosengren said. As he saw it, the Morland article would reduce the "extraordinary inventiveness" and large amounts of time and resources needed to formulate a practical design from the "vast collection of good and bad ideas and hints" in the open literature.

Similar warnings are in new affidavits filed by Robert N. Thorn, acting director of the Los Alamos Scientific Laboratory in New Mexico, and Roger E. Batzel, director of the Livermore lab.

The Morland article "is perhaps as suggestive of the process used in thermonuclear weapons as the original outline on the subject" by Edward Teller and S. M. Ulam, Thorn said.

Batzel said, "In spite of some minor technical errors, it contains or strongly suggests key concepts for the functioning of the hydrogen bomb. Once the key concepts were discovered by researchers from the United States, it took only a matter of months to translate it into practice."

In arguing for publication of the article, the Progressive, again suggesting that the 1954 law is too broad, pointed out that it could be applied to a person who communicates data already in the public domain and well knoen to scientists even of never formally declassified. The Justice Department commented:

"This objection is purely hypothetical, since the statute reaches only desclosure by a person who has reason to believe that the data will be used to harm the United States or to give assistance to other countries."