Cover Image

Not for Sale



View/Hide Left Panel

Variable Scope Search System: VS3

JACOB LEIBOWITZ, JULIUS FROME, and DON D.ANDREWS

This paper will describe a system being developed by the U.S. Patent Office for mechanizing searches for organic chemical compound disclosures. This effort is part of an overall program for mechanization of the entire searching operation of the Patent Office. Its objective is to provide a solution to the complex problems of patent searching occasioned by the rapid, exponential growth of the art, the multiple and variable points of view required in patent search requirements, and the relative inability of rigid manual classification systems to provide such multiple access to subject matter.

The usefulness of the new system is not limited to patent searching. It can be used for any search with respect to chemical compounds in terms of structural characteristics whether the search is done by the patent profession, the research scientist or the industrial organization.

Detailed descriptions of the nature of the patent search problem and some of the mechanization research toward its solution have appeared in the literature (111). This paper will give, therefore, only a brief general statement of the problem with respect to chemical compound searching for which the new structural search system is designed.

The problem

A patent search is performed with respect to the claimed subject matter of an application for a patent to determine its patentability by comparison with the prior art subject matter. The examiner searches both from the point of view of (a) novelty and (b) invention. He is therefore interested not only in identical subject matter but also in similar or the most closely related subject matter. Thus, patent searches are ordinarily from generic points of view regardless of whether the claimed subject matter is for a specific embodiment or a generic class.

JACOB LEIBOWITZ and JULIUS FROME Office of Research and Development, U.S. Patent Office, Department of Commerce, Washington, D.C.

DON D.ANDREWS Director of Research and Development, U. S. Patent Office, Department of Commerce, Washington, D. C.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 1117
--> Variable Scope Search System: VS3 JACOB LEIBOWITZ, JULIUS FROME, and DON D.ANDREWS This paper will describe a system being developed by the U.S. Patent Office for mechanizing searches for organic chemical compound disclosures. This effort is part of an overall program for mechanization of the entire searching operation of the Patent Office. Its objective is to provide a solution to the complex problems of patent searching occasioned by the rapid, exponential growth of the art, the multiple and variable points of view required in patent search requirements, and the relative inability of rigid manual classification systems to provide such multiple access to subject matter. The usefulness of the new system is not limited to patent searching. It can be used for any search with respect to chemical compounds in terms of structural characteristics whether the search is done by the patent profession, the research scientist or the industrial organization. Detailed descriptions of the nature of the patent search problem and some of the mechanization research toward its solution have appeared in the literature (1–11). This paper will give, therefore, only a brief general statement of the problem with respect to chemical compound searching for which the new structural search system is designed. The problem A patent search is performed with respect to the claimed subject matter of an application for a patent to determine its patentability by comparison with the prior art subject matter. The examiner searches both from the point of view of (a) novelty and (b) invention. He is therefore interested not only in identical subject matter but also in similar or the most closely related subject matter. Thus, patent searches are ordinarily from generic points of view regardless of whether the claimed subject matter is for a specific embodiment or a generic class. JACOB LEIBOWITZ and JULIUS FROME Office of Research and Development, U.S. Patent Office, Department of Commerce, Washington, D.C. DON D.ANDREWS Director of Research and Development, U. S. Patent Office, Department of Commerce, Washington, D. C.

OCR for page 1117
--> Another feature of patent searching is the variability in search requirements. The point of view of searching varies in accordance with the constantly shifting emphasis which is characteristic of the field of new developments and invention. Features of the new system The search system is called VS3 (variable scope search system). It is designed to provide multiple access to organic chemical compound disclosures with great variability in scope both generically and specifically. Coding of the compound for machine searching is done from the structural formula of the compound. The coding is done by nontechnical clerical personnel. The system permits the use of a simplified coding method. Each compound is regarded as consisting of certain building-block units in certain associations with each other. The building blocks are single ring configurations and selected nonring or chain-unit configurations, e.g., the benzene ring, the diazine ring, and the carboxamide chain unit. The assemblage of codes for these building blocks per se is constant in each compound, while the associations among these building blocks are variable with each compound. The constants, therefore, have preassigned codes and corresponding prepunched cards. In coding a compound, the relationships among these building blocks are indicated and the prepunched cards are assembled and completed according to the indicated associations. Another feature of the system is the fact that there is no limit imposed on the size of the dictionary providing the descriptive terminology for description of the compounds or the amount of description that can be given to the disclosure of any particular document. The system is a punched card system using the new machine ILAS previously described (1, 2). Types of search questions answered by VS3 The body of art selected for experimentation with the VS3 system is the “thiazine” art. The example in Fig. 1 constitutes the disclosure of a compound in this art, taken from U.S. Patent No. 1,996,867. FIGURE 1

OCR for page 1117
--> The system is designed to permit retrieval of this compound, inter alia, on the basis of questions of the following type. To find: An aryl methyl ether An aryl sulfonamidobenzene compound An azine A thiazine A 1,2,4-trichlorophenthiazine A 1,4-thiazine A dichlorobenzene A sulfonamidodiphenylamine An aminochlorophenothiazine and so on. It is important to note that retrieval will be, in each case, not only of the compound depicted but also of any other compounds meeting the requirements of the search. Also, the search varies in scope. It may be expressed in terms of one or several characteristics. The search may be for an ether broadly or a methyl ether more specifically, for a heterocyclic compound, or more specifically an azine, or a thiazine, or a 1,4-thiazine. It can require limitation to positions of substitution or ignore positions of substitution. Vocabulary of the system and codes A standard 80 column IBM card is used for punching the code. Codes are punched horizontally across the card. Thus there may be punched on each card 12 code words each containing 80 bits or punching positions. Figure 2 shows the format of a code word. The first 68 punching positions are divided FIGURE 2 into 17 characters of 4 punching positions each. The code is punched in hexadecimal digits. Thus this part of the word has available 17 hexadecimal digits.

OCR for page 1117
--> This is further subdivided into three fields, a field of one character for signal, a field of 4 characters (M, 1, 2, 3) for modulant and the remaining 12 characters for subject matter. The last 12 punching positions, columns 69–80, is the “interfix” field. This is further subdivided into 6 punching positions for the “homo” field and 6 punching positions for the “hetero” field. Descriptors The terms appearing in the dictionary which are used to describe and define the characteristics of the chemical compounds are designated as descriptors. The descriptors are translatable into code and so the term descriptor and code will be used synonymously. It is convenient to consider the descriptors and their code equivalents as being of two types, (1) substantive and (2) organizational. The substantive descriptors describe and define the characteristics of the chemical compounds and they appear in the coding dictionary, a portion of which is illustrated in Appendix A. (The complete dictionary is available at the Office of Research and Development, United States Patent Office). The organizational descriptors express the relationship among these characteristics. This division is purely arbitrary since substantive codes also express relationships. The division is based, however, on the means employed in the system to set forth the relationships. The substantive codes appear in the 16 characters M through 15 of the word. The organizational codes, expressive of relationships among the substantive codes, appear in the signal and interfix part of the word. Substantive codes The substantive code contains a modulant and subject matter codes. The modulant is a modifier of the subject matter codes; it is a device for using the same code to mean a variety of things according to the particular modulant used. At present, there are 12 modulants employed, although provision has been made for over 65,000 by the allotment of 4 hexadecimal characters to the modulant field. The modulants are listed at the beginning of Appendix A, and the types of information recorded in each type of word according to these modulants are briefly exemplified throughout Appendix A. Organizational codes There are two types of organization codes, (1) the grouping shown by the signal code and (2) the relationships among the groupings shown by the interfix code.

OCR for page 1117
--> GROUPING (SIGNALS) The compound is coded in terms of substructures of various sizes up to and including the complete molecule. The smallest substructural units are the ring and the chain unit. The ring is the familiar cyclic structure; the chain unit is an element or collection of elements in an acyclic configuration. The limits of the chain unit are determined by the groups appearing in the dictionary in the M-1, M-2, M-3, M-4, and M-C words. Most of these groups are the conventional “functional” groups of chemistry although there was no hesitation in synthesizing groups whenever it was deemed necessary from the point of view of retrieval. The compound of Fig. 1 has been rewritten in Fig. 3 to illustrate the grouping organizations of the structure. The structure has been transformed into the FIGURE 3 type of skeletal formula set up preparatory to coding. The rings are indicated by Roman numerals; the chain units are encircled. The next substructures to consider are the ring systems and the chains. Ring system and ring The ring system is a structural entity which comprehends within its scope one ring or a collection of rings in “fused face” relationship. There are four ring systems in the formula (Fig. 3), as follows: [(I) (II) (III)] [(IV)] [(V)] [(VI)] The parentheses are used to symbolize the enclosure of the codes pertaining to a ring, and the external brackets symbolize an enclosure of the collection of rings pertaining to the ring system. It will be noted that a ring system may con-

OCR for page 1117
--> tain only one ring as in (b), (c), and (d). Rings connected by a bond juncture are not in the same ring system, while rings in fused face relationship are. A ring system is not only a collection of codes for the individual rings it contains. It consists of other characteristics not present in the individual ring codes, such as the fused face relationship (see Appendix A) and also characteristics of the type provided for by the ring system word M-7 (Appendix A). In setting up the codes for the ring system, then, this additional information is included within the ring system grouping. Chain unit and chain The relationship between the chain unit and the chain is analogous to the relationship between the ring and the ring system. The chain is a continuity of chain units; its continuity is terminated by the interposition of a ring. It may consist of one chain unit only. The following are the chains found in the compound of Fig. 3 (with the same bracketing symbology). [(C) (O)] [(N)] [(Cl)] [(Cl)] [(Cl)] [(C)] This is again a substructure within a larger substructure. Compound grouping The next order of grouping is the enclosure of the whole set as a compound unit. Within this unit codes are then added to indicate additional information pertaining to the entire compound group. The type of information added is indicated in words M-8, M-9, M-A, M-B (see Appendix A). Patent grouping The next and final grouping is that which encloses all the compounds as pertaining to the same document. This grouping organization of the codes for a

OCR for page 1117
--> compound structure may be compared to the organization of the written language. The substantive codes constitute the identification of the letters of the alphabet and their meanings in the word; the collection of words constitutes the larger organization of the sentence. In addition, however, since the meaning of the sentence is more than the mere sum of the words, additional substantive information is added pertaining to any additional concepts provided by the sentence as a whole. The sentences then are in a group constituting the larger substructure of the paragraph, with further information pertaining to the paragraph not available as the mere contextual sum of the sentences. Signals The grouping of the codes is handled through the signals, a list of which appears in Appendix A. These signals permit the proper correlations among the codes so that the codes for one ring do not get scrambled with the codes for a different ring, or the codes for one compound do not become correlated with the codes for a different compound. There is no fixed limit to the number of codes that may be included within any signal group. Thus, any number of descriptors may be applied to definition of a ring, a ring system, a chain unit, a chain, or the compound as a whole. By the same token there is no particular limit to the amount of information that can be recorded for any one document. In the machine operation, the presence of the signal signifies the termination of all the codes pertaining to the particular structural unit defined by the signal. While properties and functions have not been encoded into the system at present, it will be obvious that this can be done according to the grouping logic (11). Properties of a compound would be grouped on the compound level; properties of a substructure would be grouped within the level of said substructure. Thus in searching, correlations can be made between compound and function or substructure and function. Interfix Another organizational relationship is that of connectivity of the substructures. In the compound of Fig. 3, ring I is joined to N, ring IV is joined to N and to ring V, ring VI is joined to chains—O—C and—NSO2—, and so on. This relationship is handled by the interfix device. This involves the assignment of a pair of identical arbitrarily selected numbers to those substructures which are joined to each other.

OCR for page 1117
--> The following types of connections are defined by interfix: (a) Ring to ring by bond juncture Homo interfix (b) Chain unit to chain unit (c) Ring to chain unit Hetero interfix (d) Ring to chain Types (a) and (b) are called the homo interfixes and are coded in columns 1 to 6; (c) and (d) are the hetero interfixes coded in columns 7 to 12. For illustration, a portion of the formula of Fig. 3 has been extracted and interfix numbers assigned as shown in Fig. 4. Each interfix number has been placed at the bond juncture to indicate that it is assigned to each of the sub- FIGURE 4 structures involved in the connection. Thus in coding, the units will have the following interfix numbers (interfix in the subscript): (C1), (O1,7), (VI7,8), (V9). The substructures of Fig. 4 can be reconstructed according to the rule that those substructures are joined to each other which have the same numbers. It must be emphasized that the value of the number is of no consequence. What is significant is the identity of the pair of numbers. Thus, the groups can be renumbered as shown in Fig. 5. The structure is FIGURE 5 equivalently defined and can be reconstructed from the interfix relationship. The interfix is punched in any of columns 69 to 80 according to the number assigned. While scanning and sensing of the codes occur row by row, determination of the interfix relationship is done by a vertical comparison down the column. Thus one code word in a row is connected to some other code word

OCR for page 1117
--> in a different row, either on the same card or on a different card, through recognition of a pair of punches in the same vertical column. This principle will be further illustrated. See Appendix B. Coding: set-subset, exact pattern Two methods of recording the codes are used according to two different desired searching devices: (1) the set-subset method and (2) the exact pattern method. To exemplify this, the code words used to describe a ring, i.e., M-5 for set-subset and M-6 for an exact pattern have been selected. In the M-5 word (see Appendix A), assignment of codes is on a bit-by-bit (set-subset) basis. In character 7, for example, the bits have been allotted the meanings shown in Fig. 6. A ring which is both unsaturated and heterocyclic will be FIGURE 6 coded as 7–5 in the M-5 word as in Fig. 7. A search for a heterocyclic ring 7–4 FIGURE 7 will involve searching for a hole in the 4 bit and will result in retrieval of the structure. Similarly, a search for an unsaturated ring 7–1 or an unsaturated heterocyclic ring 7–5 will result in retrieval of the indicated ring. Characters 9, 10, 11, 12, 13, and 14 involve the same type of coding. Thus, a search for a ring containing a heterocyclic nitrogen ortho to sulfur will result in retrieval of said structure even though other groups are present such as: In addition to the bit-by-bit basis, the terms of characters 4, 5, 6, and 8 have

OCR for page 1117
--> been coded by what may be called an “at least” basis, i.e., a ring containing 4 or more nitrogens has at least 3, 2, and 1 nitrogens; a 3 nitrogen ring contains at least 2 and 1 nitrogens, and a 2 nitrogen ring contains at least 1 nitrogen. Thus character 4 of the M-5 word will be punched as follows for a 4N, a 3N, a 2N, and a IN containing ring: FIGURE 8 The exact pattern coding is illustrated in M-6 (Appendix A). Each code is uniquely different from any other. A 4-membered ring 12–4 is not found within a 6-membered ring 12–6. Similarly, a 5 carbon ring is not found within a 6 carbon ring. In searching, the codes can be used in any combination desired. Thus, where a search is for a diazine and it is desired to exclude the 3 N-or-over containing rings, the combinations of codes requiring a 6-membered ring, at least 2 nitrogens and exactly 4 carbons will result in this exclusion. RINGS Rings are coded according to the M-5 and M-6 words as already described. In addition to the descriptors found on pages of the coding dictionary in Appendix A, a code called an index number code is assigned in characters 4, 5, and 6 of the M-6 word. A list of rings with their index numbers appears in the coding dictionary. These are unique numbers defining the specific rings as

OCR for page 1117
--> they are disclosed in the Ring Index (12). The rings in the list have already been completely coded and are maintained in a card file. In addition, a set of prepunched cards has been prepared which contain the punched codes for each of these rings. When these rings are encountered in a formula, they need not be coded again. They are merely identified and the prepunched cards are selected for further processing. As new rings are found in the disclosures they are coded, unique index numbers are assigned, and they are added to the file. The signal is coded in a separate word as the last word of the set. Thus, three codes are shown for each ring: M-5 and M-6 followed by S-1. RING SYSTEM The ring system is the sum of the codes for the rings contained within the ring system plus any other codes required to define relationships not provided for by the sum of these codes. Where there is only one ring in the ring system, two codes for the ring plus S-1 are followed by S-2, and the coding for the ring system is complete. Where there are two or more rings in the ring system, additional codes are added to define relationships not expressed by the individual ring codes. One of these relationships is the “fused face” pattern. The fused face pattern defines the relative positions of the fused faces of each ring in the system, in accordance with the graphic portrayal of the coding dictionary in Appendix A. The heavy lines refer to a position of fused face joining with another ring. The patterns are not necessarily limited to carbon rings; the same relationships are intended to be applicable regardless of the kind of ring elements. Thus the structure shows the following patterns for each of the rings:

OCR for page 1117
--> the compounds in that art is not the same as described herein. However, the VS3 system is deemed compatible with this more comprehensive system, and it will be for the future to provide a determination as to whether or not the methods should be merged. Acknowledgment Acknowledgment is made of the assistance of Mr. H.P.Luhn of IBM in helping develop some of the concepts herein described, particularly with respect to the concept of fused face patterns. Appreciation is also expressed for the assistance of Mrs. R.W.Swanson of the examining staff and Mrs. A.Replogle, Mrs. J.M.Hale, Mrs. M.Bender, and the various other members of the clerical staff in this experiment. BIBLIOGRAPHY 1. DON D.ANDREWS. Interrelated Logic Accumulating Scanner (ILAS). Patent Office Research and Development Report No. 6. Washington, 25, D. C., Department of Commerce, 1957. 2. DON D.ANDREWS, JULIUS FROME, H.R.KOLLER, JACOB LEIBOWITZ, and H. PFEFFER. Recent Advances in Patent Office Searching, Steroid Compounds and ILAS. Patent Office Research and Development Report No. 8. Washington 25, D. C., Department of Commerce, 1957. 3. DON D.ANDREWS and SIMON M.NEWMAN. Storage and Retrieval of Contents of Technical Literature, Nonchemical Information, Preliminary Report. Patent Office Research and Development Report No. 1. Washington 25, D. C., Department of Commerce, 1956. 4. JULIUS FROME and JACOB LEIBOWITZ. A Punched Card System for Searching Steroid Compounds. Patent Office Research and Development Report No. 7. Washington 25, D. C., Department of Commerce, 1957. 5. B.E.LANHAM, J.LEIBOWITZ, and H.R.KOLLER. Advances in Mechanization of Patent Searching, Chemical Field. Patent Office Research and Development Report (No. 2). Washington 25, D. C., Department of Commerce, 1956. 6. B.E.LANHAM, J.LEIBOWITZ, H.R.KOLLER, and H.PFEFFER. Organization of Chemical Disclosures for Mechanized Retrieval. Patent Office Research and Development Report No. 5. Washington 25, D. C., Department of Commerce, 1957. 7. SIMON M.NEWMAN. “Linguistics and Information Retrieval; Toward a solution of the Patent Office Problem.” Monograph Series in Linguistics and Language Studies, No. 10. Washington 25, D. C., Georgetown University Press, 1957. 8. SIMON M.NEWMAN. Linguistic Problems in Mechanization of Patent Searching. Patent Office Research and Development Report No. 9. Washington 25, D. C., Department of Commerce, 1957. 9. SIMON M.NEWMAN. Problems in Mechanizing the Search in Examining Patent

OCR for page 1117
--> Applications. Patent Office Research and Development Report No. 3. Washington, 25, D.C., Department of Commerce, 1956. 10. SIMON M.NEWMAN. Storage and Retrieval of Contents of Technical Literature, Nonchemical Information, First Supplementary Report. Patent Office Research and Development Report No. 4. Washington 25, D. C., Department of Commerce, 1957. 11. M.F.BAILEY, B.E.LANHAM, and J.LEIBOWITZ. Mechanized searching in the U.S. Patent Office. Journal of the Patent Office Society, XXXV (1953), 566–587. 12. AUSTIN M.PATTERSON and LEONARD T.CAPELL. The Ring Index, Reinhold Publishing Corp., New York, 1940. APPENDIX A Excerpts from VS3 index of codes The codes contained in each word are introduced by a modulant. Twelve modulants are presently defined as follows: Modulants M-1 to M-4 for chain units M-5 and M-6 for rings M-7 for ring systems M-8 for compounds M-9, M-A, M-B for metals M-C for carbon chain units Signals are employed to group the words pertaining to rings, ring systems, functional groups, and chains as follows: Signals S-1 pertaining to individual rings S-2 pertaining to the ring system S-3 pertaining to the chain unit S-4 pertaining to the complete chain S-5 pertaining to the complete compound S-6 pertaining to the complete patent Interfixes are employed to define the bond relationships. Twelve interfixes are provided, numbers 1–6 for homo relationships (ring to ring or chain unit to chain unit) and numbers 7–12 for hetero relationships (ring to chain). MODULANT M-1 Chain unit A B X(=Y) C D Chemical formula Name Structural formula A B X =Y C D   COF Carboxylfluoride C(=O)F Z 4–1 Z 5–1 C 6–2 =O 7–3 F 8-B Hal 9-D   CON Carboxylamide C(=O)N Z 4–1 Z 5–1 C 6–2 =O 7–3 N 8–5 Z 9–1 10–8 CON2 Isourea OC(=N)N Z 4–1 O 5–3 C 6–2 =N 7–5 N 8–5 Z 9–1 10–8 CON2 Urea NC(=O)N Z 4–1 N 5–5 C 6–2 =O 7–3 N 8–5 Z 9–1 10–8 CON3 Semicarbazide NC(=O)NN Z 4–1 N 5–5 C 6–2 =O 7–3 N 8–5 N 9–5 10–8 CO2N Urethane OC(=O)N Z 4–1 O 5–3 C 6–2 =O 7–3 N 8–5 Z 9–1 10–9 SO2I Sulfonyliodide Z 4–1 Z 5–1 S 6–4 I 8–9 Hal 9-D   SO2N Sulfonamide Z 4–1 Z 5–1 S 6–4 N 8–5 Z 9–1 10–8

OCR for page 1117
--> MODULANT M-2 Chain unit A X (=Y) B Chemical formula Name Structural formula A X =Y B CO Ketene =C(=O) =4-E C 5–2 =O 6–3 Z 7–1 CO Ketone C(=O) Z 4–1 C 5–2 =O 6–3 C 7–2 CO Quinone C=O C 4–2 C 5–2 =O 6–3 C 7–2 CS Thioketene =C(=S) =4-E C 5–2 =S 6–4 Z 7–1 CS Thioketone C(=S) Z 4–1 C 5–2 =S 6–4 C 7–2 MODULANT M-3 Chain unit A X Y Chemical formula Name Structural formula A X Y   NH Secondary amine Z 4–1 N 5–5 C 6–2   N Tertiary amine C 4–2 N 5–5 C 6–2   O Ether -O- Z 4–1 O 5–3 Z 6–1 10–4 S Thioether -S- Z 4–1 S 5–4 Z 6–1 10–4 N Quaternary amine Z 4–1 N 5–5 Z 6–1   N Imine =N =4-E N 5–5 Z 6–1 NH2 Primary amine H-N-H Z 4–1 N 5–5 H 6–6 N2 Hydrazine Z 4–1 N 5–5 N 6–5 Br Bromine -Br Hal 4-D Br 5-C Z 6–1 Cl Chlorine -Cl Hal 4-D Cl 5-A Z 6–1 F Fluorine -F Hal 4-D F 5-B Z 6–1 I Iodine -I Hal 4-D I 5–9 Z 6–1 MODULANT M-4 Chain unit Chemical formula Name Structural formula Notation code CN Nitrile −C≡N 4-B CN Isonitrile −N=C 4–6 CN2 Carbodiimide N=C=N 4–8 CNO Cyanate O−C≡N 4–3 CNO Isocyanate N=C=O 4–4 MODULANT M-C Carbon chain No. carbons No. carbons specifically No. carbons genetically CU Posn. on ring Code CU Posn. on R.S. Code   4 5 8 9   1 5–1 9–1 1 11–1 1 13–1 2 5–2 9–1 2 11–2 2 13–2 3 5–3 9–1 3 11–3 3 13–3 4 5–4 9–3 4 11–4 4 13–4 5 5–5 9–3 5 11–5 5 13–5 6 5–6 9–7 6 11–6 6 13–6 7 5–7 9–7 7 11–7 7 13–7 8 5–8 9-F 8 11–8 8 13–8 9 5–9 9-F 9 11–9 9 13–9

OCR for page 1117
--> MODULANT M-5 Ring schedule No. N atoms O atoms S atoms Misc. hetero atoms Type of ring   1 4–1 5–1 6–1 8–1 Unsaturated 7–1 2 4–3 5–3 6–3 8–3 Saturated 7–2 3 4–7 5–7 6–7 8–7 Heterocyclic 7–4 4,+ 4-F 5-F 6-F 8-F Homocyclic 7–8 Combinations of atoms Ortho Meta Para N-N 9–1 11–1 13–1 N-O 9–2 11–2 13–2 N-S 9–4 11–4 13–4 O-O 9–8 11–8 13–8 O-S 10–1 12–1 14–1 S-S 10–2 12–2 14–2 Misc. (a)-Misc. (a)a 10–4 12–4 14–4 Misc. (a)-Misc. (b)a 10–8 12–8 14–8 a Miscellaneous combinations include all combinations not specifically provided for. Misc. (a)-Misc. (a) refers to combinations of the same elements. Misc.(a)-Misc.(b) refers to combinations of different elements. MODULANT M-6 Ring schedule No. elements in ring Code No. carbons in ring Code 1   1   2   2   3 12–3 3 13–3 4 12–4 4 13–4 5 12–5 5 13–5 6 12–6 6 13–6 7 12–7 7 13–7 8 12–8 8 13–8 9 12–9 9 13–9 10 12-A 10 13-A 11 12-B 11 13-B 12 12-C 12 13-C 13 12-D 13 13-D 14 12-E 14 13-E 15,+ 12-F 15,+ 13-F

OCR for page 1117
--> MODULANT M-6 Ring schedule MODULANT M-7 Ring system schedule No. Rings N rings O rings S rings Benzene rings Miscellaneous hetero rings 1 7–1 8–1 9–1 10–1 11–1 12–1 2 7–2 8–2 9–2 10–2 11–2 12–2 3 7–3 8–3 9–3 10–3 11–3 12–3 4 7–4 8–4 9–4 10–4 11–4 12–4 5 7–5 8–5 9–5 10–5 11–5 12–5

OCR for page 1117
--> MODULANT M-8 Compound schedule No. Ring systems Saturated rings Unsatd. rings 7 and 7+ rings N rings O rings S rings Benzene rings 1 4–1 5–1 6–1 7–1 8–1 9–1 10–1 11–1 2 4–2 5–2 6–2 7–2 8–2 9–2 10–2 11–2 3 4–3 5–3 6–3 7–3 8–3 9–3 10–3 11–3 4 4–4 5–4 6–4 7–4 8–4 9–4 10–4 11–4 5 4–5 5–5 6–5 7–5 8–5 9–5 10–5 11–5   etc.   MODULANTS M-9, M-A, M-B Metals schedule Symbol Metal M-9 M-A M-B     M-B Ca Calcium 8–1   5–1   Group IA 4–1 Cr Chromium   4–1 6–1 10–1 Group IIA 5–1 Co Cobalt   6–1 6–1 12–1 Heavy metal 6–1 Cu Copper 13–1   6–1 13–1 Group IIIB 7–1 Au Gold 13–1   6–1 13–1 Group IVB 8–1 Fe Iron 10–1   6–1 12–1 Group VA 9–1 Pb Lead   11–1 6–1 15–1 Group VIB 10–1 Li Lithium 11–1   4–1   Group VIIB 11–1 etc.  

OCR for page 1117
--> APPENDIX B FIGURE B-1. Preparatory skeletal formula.

OCR for page 1117
--> FIGURE B-2

OCR for page 1117
--> FIGURE B-3

OCR for page 1117
--> FIG. B-4. Code meanings for phenothiazine

OCR for page 1117
--> FIGURE B-5. Outline of the punched card used in VS3. The twelve codes reading from row 9 up correspond respectively to the first twelve codes shown in Fig. B-3