2010 volume 7 number 1 - issiran.com

2010 VOLUME 7 NUMBER 1

Iron & Steel Society of IranIsfahan University of Technology

International Journal of

Published each six months by Iron and Steel Society of Iran with collaboration of Isfahan University of Technology

International Journal of Iron & Steel Society of Iran (Int. J. of ISSI)

Managing Editor: Prof. A. Najafizadeh (Isfahan University of Technology, Iran)

Editor: Prof. A. Saidi (Isfahan University of Technology, Iran)

Editorial Board Prof. A. Abbaschian (Materials Eng. Dept., University of Florida, USA)Prof. B. Bavarian (Materials Eng. Dept., University of California, USA)Prof. J. H. Beynon (Faculty of Engineering & Industrial Sciences, Swinburne University of Technology, Australia)Prof. T. Chandra (Materials Eng. Dept., University of Wollongong, Australia)Prof. A. J. DeArdo (Materials Eng. Dept., University of Pittsburgh, USA)Prof. I. Garcia (Materials Eng. Dept., University of Pittsburgh, USA)Prof. H. Henein (Materials Eng. Dept., University of Alberta, Canada)Prof. P. Hodgson (Materials Eng. Dept., Deakin University, Australia)Prof. M. Jahazi (Mining Metals and Materials Eng. Dept., Iran University of Science of Technology, Iran)Prof. S. A. Jenabali Jahromi (Materials Eng. Dept., Shiraz University, Iran)Prof. J. J. Jonas (Mining Metals and Materials Eng. Dept., McGill University, Canada)

Prof. N. Kanani (Materials Science Department, Technical University of Berlin, Germany)Prof. A. Monshi (Materials Eng. Dept., Isfahan University of Technology, Iran)Prof. M. Moshksar (Materials Eng. Dept., Shiraz University, Iran)Prof. A. Najafizadeh (Materials Eng. Dept., Isfahan University of Technology, Iran)Prof. R. A. Rapp (Materials Eng. Dept., Ohio State University, USA)Prof. A. Saatchi (Materials Eng. Dept., Isfahan University of Technology, Iran) Prof. A. Saidi (Materials Eng. Dept., Isfahan University of Technology, Iran)Prof. T. Sakai (Mechanical Engineering & Intelligent Systems Dept., University of Electro Communications, Japan)Prof. R. W. Waterhouse (Materials Eng. Dept., University of Nottingham, UK)Prof. J. Wood (Materials Eng. Dept., University of Nottingham, UK)Prof. H. Yoozbashizadeh (Materials Science and Eng. Dept., Sharif University of Technology, Iran)Dr. H. Edris (Materials Eng. Dept., Isfahan University of Technology, Iran)Dr. E. Keshavarz Alamdari (Mining & Metallurgical Eng. Dept., Amirkabir University of Technology, Iran)Dr. A. Shafyei (Materials Eng. Dept., Isfahan University of Technology, Iran)Dr. M. Shamanian (Materials Eng. Dept., Isfahan University of Technology, Iran)

Executive Editorial Board: Y. Mazaheri, N.Orak shirani Publisher: Iron & Steel Society of Iran

International Journal of Iron and Steel Society of Iran is published each six month by Iron and Steel Society of Iran with collaboration of Isfahan University of Technology. Annual subscription rates are as follows: for university departments, libraries, industrial, other multiple-reader institutions and individuals in Iran 100000 Rials. In other countries 200 US Dollars (including packing and postage). Reprints articles are available for 25000 Rials per copy in Iran and 50 US Dollars per copy else where (including packing and postage).

Iron & Steel Society of IranScience and Technology Sheikh Bahai Park, Isfahan Science and Technology Town,

Isfahan University of Technology Boulevard, Isfahan, 84156-83111, Iran.TeleFax: +98 311 3932121-24 www.issiran.com E-mail: [email protected]

International Journal of Iron & Steel Society of Iran

Referees for This Issue

Prof. A. Saatchi (Materials Eng. Dept., Isfahan University of Technology, Iran)Prof. M.H. Fathi (Materials Eng. Dept., Isfahan University of Technology, Iran)Prof. A. Saidi (Materials Eng. Dept., Isfahan University of Technology, Iran)Prof. A. Najafizadeh (Materials Eng. Dept., Isfahan University of Technology, Iran)Dr. H. Edris (Materials Eng. Dept., Isfahan University of Technology, Iran)Dr. M. Shamanian (Materials Eng. Dept., Isfahan University of Technology, Iran)Dr. A. Shahandeh (Industrial and Systems Eng. Dept., Isfahan University of Technology, Iran)Dr. M. Abzari (Management Dept., University of Isfahan, Iran)Dr. A. Etebarian (Public Management Faculty, Islamic Azad University, Khorasgan Branch, Isfahan, Iran)Mr. M.H. Joulazadeh (Chairman of Ajineh Gostar Espadana Co., Isfahan, Iran)

International Journal of ISSI, Vol.7 (2010), No.1, pp.1-10

An Analytical model of the organizational culture evaluation in Iran steel industry: a survey research of Tuka Steel Investment

Holding Company

A. Gholizadeh1* , R. Ebrahimzadeh2

Cultural management department, Islamic Azad University Khorasgan(Esfahan) Branch, Esfahan, Iran

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------AbstractThe present study has objectively investigated and reviewed 164 managers, employees and workers of Tuka Steel Investment Holding Company (TSIH Co.) as one of the most successful companies in Iran's steel industry. The Survey approach has been adopted to serve the following purposes of this study: a) to define the prevailing organizational culture in steel industry; and b) To review alignment dominant in the prevailing organizational culture with emphasis on the TSIH Co. To achieve the above-mentioned objectives, a researcher- made questionnaire was developed according to Freeman and Cameron model of organizational culture to investigate the dominant culture and the aligenment of the organizational culture. The results of data analysis using ANOVA with repeated measures showed that the dominant organizational culture in these organizations was hierarchical. On the other hand, there was lack of alignment between the dimensions of organizational culture. The research findings showed that due to the governmental structure of organizations, lack of competitiveness, conflicting views of managers in these organizations, appointments of the managers on the basis of connections and lack of stability and cohesion within the active organizations of this industry, they have faced difficulties, the ultimate outcome of which is the lack of conformity and alignment in organizational culture. The researchers, by presenting the research results, intended to find an appropriate approach and orientation to assess organizational culture in Iran steel industry with emphasis on Tuka Steel Investment holding (TSIH CO) in order to present suitable strategies to strengthen or improve the above-mentioned conditions.

Keywords: organizational culture, cultural aligenment, organizational culture models, steel industry, Iran.-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

* Corresponding author:Tel: +98 (311) 6692468, Fax: +98 (311) 5354060 E-mail: [email protected]: Cultural management department, Islamic Azad,University Khorasgan(Esfahan) Branch, Esfahan, Iran1,2. Assistant Professor

1. Introduction The word and concept of culture have been the basic studies of anthropology and sociology for more than a century, and They have been looked upon from different perspectives in these scientific fields. Researchers have established a large amount of inquiries, discussions and investigations in this field and founded a considerable scientific basis as inputs for the interdisciplinary fields in all areas of social sciences. In the decades of 1940 and 1950, most of these research studies, such as “Ralph Linton”, “Ruth Benedict” and “Margaret Mead” have dealt with the customs and traditions dominant in societies, especially primitive societies, and then tried to extract the same concepts in industrial societies. This trend has also been partly observed among sociologists; they also extracted customs in the workplace and discussed these factors within the framework of the work culture and factory. The recent survey, though, shows the formation of early written texts in the years of late

1960 and early 1970 1). The concept of organizational culture has been derived from the term culture which in terms of terminology has many concepts and meanings and, with regard to different approaches, takes up a special meaning (Fig. 1) 2). This makes it difficult for us to provide a single definition of organizational culture. In this study, we take organizational culture as a system of joint inferences that members of the organization have of an organization, and this feature leads to the separation of the two organizations from each other 3).Considering this definition of organizational culture, Furnham & Gunter divided the duties and functions of organizational culture into two parts: 1. To consolidate the processes of the organization and socialization of the staff, leading to a sense of identity and character building for staff, and their commitment to the organization. 2. Coordination within the organization to create competitive advantages for the organizations, environmental organizations understanding, and stability and harmony in the social system of the organization 4).Review of the research literature on organizational culture reveals that the role and importance of organizational culture had first been put in the form

International Journal of ISSI, Vol.7 (2010), No.1

2

of more general issues like the effectiveness of the organization by 5-7) Peters & Waterman (1982), Deal & Kennedy (1982), and then Kotter & Hesket (1992) developed this concept by emphasizing the importance of strategic fit between the organization and its environment and the need for the quality compliance of the organization. There is little agreement among researchers concerning how to study and understand organizational culture. This is due to the type of methodology used by the researchers. Some theorists argue that in cultural studies, some quantitative methods, are simplistic and low in values 8) and can not identify the assumptions and values governing an organization 9,10).In contrast, some other researchers believe that cultural researchers should avoid methods in anthropology and ethnography and go beyond the exploratory approach to develop a framework that enables comparison and matching 11-12). Recent researches have examined the organizational culture and its effects on the organization with better orientations. Some researchers have evaluated the effects of culture on the quality of products and services and come to the conclusion that in cultures where individualism, masculinity and long-term goals exist, and there is a hierarchical organizational culture, more attention is placed on quality management of services and products 13-15).There are other beliefs that culture has a dramatic effect on leadership style, and through this it will affect power, control, reward systems and decision-making processes 16).Other research results also indicate that culture influences organizational commitment and job satisfaction. This issue is particularly observed in more innovative and

supportive cultures 17-19). In addition, reviews show that culture strongly influences attitude, behavior, and organization management changes, and improve the effectiveness of organizational performance 20-23). Citing the literature on this research, the present paper is trying to use an appropriate model of organizational culture to examine and investigate TSIH Co, and its subsets as a highly successful organization in the steel industry of Iran, in order to assess the strengths and weaknesses of the organizational culture. 2. Review of the models of organizational cultureIn the organization and management literature, organizational culture researchers and theorists have tried to develop and design conceptual frameworks, models or measurement tools to identify specific organizational culture. Some examples of these models are: - Denison et al. Model of organizational culture 23).- Hatch cultural dynamics model 24).- Goffee and Jones Model of organizational culture 17).- Freeman and Cameron model of organizational culture 25).- Bath Model of organizational culture 26).- Geert Hofstede Model of organizational culture patterns 12).- Schein model of organizational culture 10).With the investigation of different types of models, this study has chosen the pattern of organizational culture of Cameron and Freeman for the following reasons: 1. Using this model, it is possible to compare different organizational cultures, and a framework can be found to evaluate organizational culture in the steel industry with an emphasis on Tuka Steel Investment Holding

Fig. 1.Organizational culture terminology 3).


3

Company (TSIH Co.).2. Each of the proposed cultures in this model has its own dominant characteristics, leadership style, focus, values and strategic orientations. This issue allows researchers to acquire a more comprehensive assessment of the organizational culture, while other mentioned models do not have such a capability. 3. Using the above model, it is possible to analyze cultural alignment. The concept of cultural alignment refers to the compatibility of different aspects of culture. Research27) has shown that although cultural alignment is not a prerequisite of success of the organization, it is an essential pattern of successful organizations.4. The existence of nonalignment in the organization is often taken as a warning for the need to change in the organization, and only this model helps the researchers to recognize the need to change.In sum, Freeman and Cameron model strives to provide an appropriate research framework to evaluate organizational culture. The framework has been based on four sets of characteristics:1.The dominant characteristic or values2.The dominant leadership style in the organization3.The requirements and commitments “primary focus”4.The organization’s current strategic emphasis. This model form is shown in Fig. 2.As Fig. 2 shows the vertical axis represents the spectrum of dynamic processes rather than static and mechanical processes, the focus of which ranges from flexible, spontaneous and natural to static and control, etc. The horizontal axis on the one hand maintains the stability and processes within an organization (integrity,

attention to simplicity and repetitive activities) and on the other hand, emphasizes the situations and external conditions (competition, differentiation and etc.). Thus, the result includes four types of organizational culture: 1. Clan culture. 2. Adhocracy culture3. Hierarchical culture. 4. Market Culture.Each of the cultures discussed above has its own leadership style, focus, values and strategic focus25). With regard to the researcher’s desired model and research pattern, this paper tries to offer a framework in order to assess the organizational culture in Iran’s steel industry with an emphasis on TSIH Co. through the evaluation of four different types of organizational culture model of Cameron and Freeman. Since the steel industry in Iran has not taken any specific evaluation to recognize organizational culture and there is no certain framework either, so the questions were considered as follows:a) What is the dominant organizational culture in TSIH Co. according to the four types of organizational culture (clan, adhocracy, hierarchy and market)? b) Are the dominant dimensions of organizational culture in TSIH Co. aligned together? Thus, by answering these questions, it can be expected to identify the organizational culture in Iran’s steel industry with an emphasis on TSIH Co. and by examining the dominant characteristics, leadership style, focus and strategic orientation and then it is possible to present a comprehensive view of organizational culture in Iranian steel industry in order to compare it with

Fig. 2. Organizational Culture Model of Cameron and Freeman25).


4

background of research and provide a better evaluation for the organizational culture. 3. Method3.1. participantsThe participants in this study were164 people (9 females and 155 males) from the workers and the staff of TSIH Co.. Based on survey research method and using a statistical formula of limited samplings the study began. Participants’ age ranged from under 25 to 55 and their work experience ranged from under 5 years to 15 years. The information about the participants in this study is shown in table 1.The reasons for choosing the participants of this study from TSIH Co. in Iran had the following characteristics: 1. The company is a subsidiary company of Mobarakeh Steel Co., the largest and the most equipped steel company in Iran. The review of the company’s organizational culture can also lead to a more complete understanding of the organizational culture of the country’s steel industry organizations which are active in this field. 2. TSIH Co. with 25 subsets is active in various fields of steel industry. Extensive activities of these companies from design, production and transport of steel products in the country has led us to review this holding to find a more accurate and complete cultural organization in the steel industry in Iran.3. TSIH Co. has been one of the most successful companies concerning performance during the past 10 years in Iran. So, the review of the company’s organizational culture helps us in the analysis of the organizational culture.Procedure All the participants in this study completed the organizational culture questionnaire which was designed on the basis of Cameron model and were provided with necessary directions by the researcher at the posts. Thus, 200 organizational culture questionnaires were distributed among participants and after collecting them during a two-month period in 2010, 164 questionnaires from the distributed questionnaires were delivered to researchers. The return rate of the questionnaires was 82 percent which, according to data analysis, is considered a good return rate.

3.2 Measures The measurement tool in this study is the organizational culture questionnaire based on the model of Cameron and Freeman with two parts: the first part of questionnaire included five questions that dealt with the demographic variables (gender, years of service, age and organizational posts); the second part included 24 questions aimed at measuring four types of organizational culture (clan, hierarchy, market and adhocracy). In designing this section of the questionnaire, five-option Likert spectrum was used. After the necessary reforms and the assessment of the

validity and reliability of the tool, the questionnaire was turned into 16 standard questions. Thus, the first 4 questions of the questionnaire deal with clan measures of the organizational culture. Below is an example of these questions to be presented: “How sincerely close is your relationship with each other, and do the people in your organization see each other as a clan?”

Table 1. Conditions in the steel industry. Demographic status of respondents

l TOTAL Percent Age of the respondentsUnder 25 6 3.725-30 84 51.235-45 57 34.845-55 17 10.3Gender of the respondentsMen 155 94.5Woman 9 5.5Education level of the respondentsBelow diploma 18 11Diploma 67 4.8AA 31 18.9BA and BS 42 25.6MA and PhD 6 3.7Job and position of the respondentsWorker 74 45.1Staff and experts 74 45.1 Senior experts and managers 16 9.8Employment history of the respondentsBelow 5 years 29 17.75-10 56 34.1 10-15 50 30.5Above 15 years 29 17.7

The next four items (questions 5 to 8) consider adhocracy culture, an example of these items is: “How does your organization pay attention to innovations and developments of new products and stresses on being the best in the market?” The set of four other questions (questions 9 to 12) is related to hierarchy organizational culture, an example of these items can be such as: «To what extent does your organization emphasize the stability, efficiency and maintain the current state of organization as their, focuses?» The 4 last items of the questionnaire (questions 13 to 16) measure market Organizational Culture, a sample of these items is: «How does your organization emphasize goals and activities and does it pay attention to the substantial production in the organization?» The respondents were requested to rate their agreement according to any one of the items in a range of five options (very low to very high). At first the reliability rate of the four sections for each organizational culture (hierarchy, market, clan and adhocracy) was measured and the total amount of reliability of the questionnaire was calculated. The results taken from


5

Cronbach>s Alpha method showed good reliability for the measurement tool. The results of the reliability measurement are shown in the table below. Content validity of the organizational culture questionnaire was examined by the university professors, scholars and other researchers in Iran; and after applying the reforms and changes required by the experts, it was approved.After confirming the content validity, construct validity of the questionnaire was determined by using exploratory factor analysis with the principal component approach and orthogonal rotation method and varimax. Thus, 8 items were deleted from 24 questions and only 16 questions remained that measured the four types of organizational culture in Cameron and Freeman model (Table 3).

Table 2. Reliability of measurement tool and its dimensions.

Standard value ReliabilitycoefficientVariables

More than 0.70.721Clan culture

More than 0.70.886Adhocracy

More than 0.70.752Hierarchy culture

More than 0.70.823Market culture

More than 0.70.938 Reliability coefficient

calculated for the wholeinstrument measure

Then, the researchers conducted a second order confirmatory factor analysis using the remaining 16

items from the exploratory factor analysis. Second order factor analysis model of organizational culture questionnaire is shown with the following factor loadings. In addition, a summary of goodness of fit indexes of the model is shown in Table 4.

Fig. 3. Second order factor analysis model of organizational culture and its dimensions.

Considering all the suitable indicators fitting the model, it can be concluded that second order factor

Table 3. Results of exploratory factor analysis of the questionnaire of the organizational culture.

Component Hierarchy Market Clan Adhocracy

Family relationship -.144 .120 .718 .072Mentor leadership style .147 .093 .673 .080Commitment and loyalty to organization .157 .024 .631 .086Human resources focus and group cohesion .078 .152 .723 -.017Risk taking and innovator leadership style .323 -.007 .078 .608Dynamic and entrepreneurial organization -.135 .163 -.019 .760Commitment to innovation and development .305 .052 .120 .672Focus on growth and acquiring new resources -.051 .092 .074 .684Formalized and structured organization .613 .169 .061 .225Manager as coordinator and organizer .646 .091 .078 .025Organization with formal rules and policy .730 .037 .146 -.005Focus on permanence and stability .627 .095 -.028 .028Production oriented organization .039 .700 .189 .035Manager as producer and technician .044 .697 .093 .151Organization with task and goal accomplishment .242 .591 .060 .063Focus on measurable goals

Eigen valuesPercentage of variance

.088

3.42612.993

.741

1.67612.53

.053

1.55012.48

.049

1.39712.31

Extraction Method: Principal Component Analysis.Rotation Method: Varimax with Kaiser Normalization.


6

analysis model of organizational culture questionnaire is a good fit. Another important point is the construct reliability assessment of the four organizational cultures (hierarchy, market, clan and Adhocracy). Construct reliability for each considered dimension has been calculated as follows 28):

Structural reliability =jloadingstd

loadingstdε∑+∑

∑2

2

).().(

jε : is the measurement error Reliability Coefficients of the four organizational culture ranges from 0.70 to 0.77 which corresponds to the standard value of 0.70 submitted by 23). Thus, findings from the reliability structure are desirable(Table 5).

4. Results of the study The data analysis in this study was conducted to investigate the following two questions of the research: (a) the identification of the dominant organizational culture with regard to the review survey of Iran’s steel Tuka Steel Holding Investment; (b) the review and survey of the existence of alignment among fundamental aspects of the dominant organizational culture in these organizations based on Cameron and Freeman model (alignment between the dominant features of organizational culture, leadership style, focus and strategic emphasis). Table 6 shows the Descriptive statistics of the four types of culture. At first, the mean differences were analyzed between the four types of organizational culture (clan, Adhocracy,

hierarchical, and market) to determine which form of the four organizational cultures is dominant in Iran’s active steel factories. One-way repeated measures analysis of variance (ANOVA) was used to compare related data with four types of organizational culture. The findings of Table 7 indicate significant differences between the four types of organizational cultures (P = 0.011, F (3,489)=3.734). The results of pair-wise comparisons which are related to the dominant organizational culture in Tuka Steel which is presented in Table 8 help us compare the four types of organizational culture to determine the dominant organizational culture in TSIH CO.

Table 6. Descriptive statistics related to different types of organizational culture.

Descriptive Statistics Mean Std. Deviation N

Clan 2.8552 .77982 164Adhocracy 2.8613 .80938 164

Hierarchy 3.0107 .82724 164

Marke 2.9040 .78882 164

The results of the pair-wise comparisons show that hierarchical culture is significantly different from other organizational cultures; so, it can be asserted that the dominant organizational culture in TSIH CO is hierarchical.

Table 4. Goodness of fit index model.

conclusion Standard amount ofthe index

The amount of index in theappointed model Index

Model fitness is suitable-132.792χModel fitness is suitableMore than 0.050.09639 P-ValueModel fitness is suitableMore than 0.90.92GFIModel fitness is suitableMore than 0.90.90AGFIModel fitness is suitableMore than 0.90.95NFIModel fitness is suitableMore than 0.90.98CFIModel fitness is suitableLess than 0.10.049RMSEA

Table 5. Evaluation of the existing reliability factors in the model of organizational culture.

conclusionStandard valueObserved valueThe surveyed factor

Reliability is suitableMore than or equal to 0.70.70Clan culture

Reliability is suitableMore than or equal to 0.70.77Adhocracy culture

Reliability is suitableMore than or equal to 0.70.77Hierarchical culture

Reliability is suitableMore than or equal to 0.70.72Market culture


7

Table 7. ANOVA results table with repeated measurements associated with different types of the

organizational culture.Tests of Within-Subjects Effects

culture Type III Sum ofSquares

df Mean Square F Sig.

within-subjectsculture 2.547 3 .849 3.734 .011

Error(culture) 111.172 489 .227

The second section examines and analyses the alignment between the fundamental aspects of the dominant organizational culture of TSIH CO, to determine whether the four dimensions of organizational culture put forward by Cameron and Freeman (dominant characteristics of organizational culture, leadership style, focus and strategic emphasis) are in alignment with the dominant organizational culture ( hierarchy culture)?The results obtained from the one-way repeated measure ANOVA showed significant differences between the dominant characteristics of four types of organizational culture. (P =0.001, F (3,489) =5.98)

Table 9. ANOVA results table with repeated measurements associated with different dominant

characteristics in four types of organizational culture.

Tests of Within-Subjects Effects

Source Type IIISum of

Squaresdf Mean Square F Sig.

Dominantcharacteristics 15.091 3 5.030 5.979 .001

Error(Dominant) 411.409 489 .841

The pair-wise comparison table shows that the dominant organizational culture characteristics of the Tuka Company are in alignment with the dominant characteristics of the market and adhocracy. So, there

is no alignment between the dominant features in TSIH CO, and its organizational Culture (hierarchy culture). The results relating to leadership style confirm that there is a difference in mean between the one way analysis of variance with repeated measures concerning the leadership style. (P =0.006, F (3,489) = 4.24).The Table related to pair-wise comparisons about leadership style shows the dominant leadership style in TSIH CO, leadership style is closer to hierarchy

organizational culture from other leadership organizational cultures. So alignment exists between the dominant organizational culture of TSIH CO and leadership style.

Table 10. Names of variables used in pair-wise comparison.

Within-Subjects Factors Dominantcharacteristics Dependent Variable

1 Family relationship2 Creativity and adaptability3 Rules and regulation4 Competition and attainment of goals

Finally, the paired comparisons revealed no significant difference in focus; therefore, it can be said that TSIH CO does not have a certain focus and there is no alignment between Tuka Steel hierarchical culture and the focus of the organizational culture.

The results of one-way analysis of variance with repetitive values related to strategic orientation shows a significant difference and the paired comparison table shows that the strategic emphasis of the Tuka Steel Investment Holding is in alignment with hierarchy organizational culture.

Table 8. Pair-wise Comparisons.

(I) culture (J) culture Mean Difference(I-J) Std. Error Sig.a 95% Confidence Interval for Differencea

Lower Bound Upper Bound

Clan Adhocracy -.006 .050 .902 -.104 .092Hierarchy -.155* .057 .007 -.268 -.043

Market -.049 .057 .397 -.162 .065

Adhocracy Clan .006 .050 .902 -.092 .104

Hierarchy -.149* .051 .004 -.250 -.049Market -.043 .051 .407 -.144 .059

HierarchyClan .155* .057 .007 .043 .268

Adhocracy .149* .051 .004 .049 .250Market .107* .049 .030 .010 .203

MarketClan .049 .057 .397 -.065 .162

Adhocracy .043 .051 .407 -.059 .144Hierarchy -.107* .049 .030 -.203 -.010


8

Table 12. ANOVA results table with repeated measurements associated with different leadership

styles in four types of organizational culture.


Source Type III Sum ofSquares

df MeanSquare F Sig.

leadership 11.439 3 3.813 4.242 .006Error(leadership) 439.561 489 .899

Table 15: ANOVA results table with repeated measurements associated with different focus in four

types of organizational culture.Tests of Within-Subjects Effects


df MeanSquare F Sig.

Focus 2.323 3 .774 1.159 .325

Error(focus) 326.677 489 .668


Within-Subjects Factorsleadership Dependent Variable

1 Mentor leadership style2 Risk taker and innovation in leadership style3 Manager as coordinator and organizer4 Manager as technician


Within-Subjects Factors

Focus Dependent Variable

1 Loyalty, traditional cohesion within groups

2 Risk taking and flexibility

3 Rules and regulation

4 Competition

Table 11. Pair-wise Comparisons in dominant characteristics.

(I)

Orientation

(J)

Orientation

Mean Difference

(I-J)Std. Error Sig.a

95% Confidence Interval forDifferencea


12 -.110 .095 .250 -.298 .0783 .159 .092 .086 -.023 .3404 -.256* .103 .014 -.459 -.053

21 .110 .095 .250 -.078 .2983 .268* .095 .005 .081 .4554 -.146 .108 .176 -.359 .066

31 -.159 .092 .086 -.340 .0232 -.268* .095 .005 -.455 -.0814 -.415* .113 .000 -.639 -.191

41 .256* .103 .014 .053 .4592 .146 .108 .176 -.066 .3593 .415* .113 .000 .191 .639

Table14. Pair-wise Comparisons in leadership styles.

(I) leadership (J) leadership Mean Difference(I-J) Std. Error Sig.a

95% Confidence Interval forDifferencea


12 .220 .117 .062 -.011 .4503 -.134 .130 .303 -.391 .1224 -.061 .131 .642 -.320 .198

21 -.220 .117 .062 -.450 .0113 -.354* .075 .000 -.501 -.2064 -.280* .081 .001 -.441 -.120

31 .134 .130 .303 -.122 .3912 .354* .075 .000 .206 .5014 .073 .077 .341 -.078 .225

41 .061 .131 .642 -.198 .3202 .280* .081 .001 .120 .4413 -.073 .077 .341 -.225 .078


9

Table 18. ANOVA results table with repeated measurements associated with different strategic

orientation in four types of organizational culture.



df Mean Square F Sig.

Strategic 68.201 3 22.734 31.333 .000Error(strategic) 354.799 489 .726

In short, the analysis shows that the dominant organizational culture in Tuka Steel is hierarchical and it conforms to the leadership style dimensions


Strategic Dependent Variable

1 Commitment and ethics

2 Pay attention to creativity, human growth

3 Stability

4 pay attention to market and competition

and strategic orientations, but the dimensions of the dominant characteristics and focus are not significantly in aligned with the Tuka steel hierarchical culture.

Table 17. Pair-wise Comparisons in focus dimension.

(I) focus (J) focus Mean Difference (I-J)

Std. Error Sig.a

95% Confidence Interval for Differencea


12 .037 .093 .695 -.147 .2203 .140 .097 .149 -.051 .3314 .128 .090 .157 -.050 .306

21 -.037 .093 .695 -.220 .1473 .104 .095 .275 -.083 .2904 .091 .083 .271 -.072 .255

31 -.140 .097 .149 -.331 .0512 -.104 .095 .275 -.290 .0834 -.012 .083 .884 -.177 .153

41 -.128 .090 .157 -.306 .0502 -.091 .083 .271 -.255 .0723 .012 .083 .884 -.153 .177

Table20. Pair-wise Comparisons in strategic orientations.

(I) strategic (J) strategic Mean Difference (I-J) Std. Error Sig

95% Confidence Interval for Difference


12 -.171 .085 .047 -.339 -.0023 -.787 .106 .000 -.996 -.5784 -.006 .103 .953 -.210 .198

21 .171 .085 .047 .002 .3393 -.616 .090 .000 -.794 -.4374 .165 .082 .047 .002 .327

31 .787 .106 .000 .578 .9962 .616 .090 .000 .437 .7944 .780 .095 .000 .593 .968

41 .006 .103 .953 -.198 .2102 -.165 .082 .047 -.327 -.0023 -.780 .095 .000 -.968 -.593


10

5.Conclusion According to the studies conducted by researchers from observations in connection with the implementation of total quality management in TSIH CO the former correct research results are confirmed because this organization has been able to implement TQM effectively. On the other hand, cultural nonalignment in research allows researchers to conform to examine cultural alignment. The research done by27) has shown that although cultural alignment is not a precondition to the success of the organization; but it is a model of successful organizations and it is essential. Review27) of the research results shows that in TSIH CO, there is no alignment between hierarchy dimensions in the company culture. According to Quinn and Cameron view, the organizations that do not have cultural alignment need change. On this basis and with regard to the Tuka Steel Holding conditions in particular, and organizations active in the steel industry in general, they all need to have major developments and changes. This problem (need to change) is felt with regard to the conditions in the entire organizations in Iran; the issue is the outcome of social, economic, cultural conditions and... One of the most important reasons for the lack of cultural alignment in size and the dominant characteristics and focus in Tuka in particular and in other steel industries in general is that state government agencies rule them, and this makes the growth of these organizations slow. Lack of proper competition from competitors, lack of proper organization in activities, quite different and conflicting views of managers, completely personalized selection of the managers and lack of stability, alignment and cohesion in the affairs of such organizations, are considered as other problems and limitations of such organizations in Iran. On the other hand, since managers in these organizations change rapidly and cannot have the necessary job stability; therefore, the implementation of their ideas cannot be fully achieved in organizations, because every manager has his own views and new approach to Organization. This leads to instability in the organization and will bring nonalignment in the culture of the organization.

AcknowledgementsTo design and produce this articles, a large number of managers, experts and workers at Tuka Steel Investments Holding (TSIH Co.) played a significant and considerable role. Therefore, the researchers appreciate the good work they did, and thank them for their sincere cooperation.

References[1] A. Farhangi: Management knowledge, 55 (2001), 24.[2] Y. Allairre and M. E. Firsirotu: Organ. stud. 5(1984), 193.[3] S. P. Robbins: Organizational behavior, Prentice

Hall (2007).[4] P. Martins and S. Terblanche: Eur. j. innovat.manag. 1(2003), 64.[5] R. Garg, and J. Ma: Benchmarking: an international journal, 3 (2005), 260-274.[6] D. Denison and J. Janovics: The international institute for management development and university of Michigan business school (2006).[7] J. Bodley: Mayfield A fascinating introductory to anthropology(1996). [8] W. Ouchi and A.Wilkins: Annual Review of Anthropology, 11 (1985), 457.[9] C. C. Lundberg: On The Feasibility of Cultural Interventions in Organizations, Organizational Culture (1985), 169.[10] E. H. Schein: Organizational Culture and Leadership. San Francisco: Jossey-Bass(1985).[11] G. Hofstede, B. Neuijen, D. D. Ohayv, G. Sanders: A Qualitative and Quantitative Study across Twenty Cases, Administrative Science Quarterly(1990).[12] G. Hofstede: J. Manage. Stud. 3 (1986), 253.[13] M. Kumar and S. Sankaran: The TQM magazine, 2(2007),176.[14] D. Prajogo and C. McDermott: Int. j. oper.prod. man. 11 (2005), 1101.[15] A. Kanousi: Managing service quality,1(2005), 57.[16] A. Shahin and P. Wright: The leadership & organization development journal, 6 (2004), 499.[17] Z. Abdul Rashid, B. Abdul Rahman ,F. Sambasivan.:The leadership & organization development journal, 2 (2004), 161.[18] P. Lok and J. Crawford: Journal of management development, 4(2004), 321.[19] C. Silverthorne: The leadership & organizational development journal, 3 (2004), 25.[20] A. Liu, Z. Shuibo, L. Meiyung: Engineering construction and architecture management, 4 (2006), 327.[21] A. Chandrakumara and P. Sparrow: Int. j. Manpower, 6(2004), 564.[22] J. Hoogervorst: Journal of managerial psychology, 3(2004), 288.[23] C. Fey, D. Denison: SSE/EFI working paper series in business administration (2000).[24] M. J. Hatch: Organization Theory Modern, symbolic and postmodern perspectives, Oxford University Press (2006)2. [25] D. Lund: Journal of business & industrial marketing, 3(2003), 219.[26] P. Hawkins: Human relations, 4(1997).[27] K. S. Cameron and R. E. Quinn: Diagnosing and Changing Organizational Culture, Reading: Addison Wesley (2006).[28] C. Fornell and D. F. Larcker: J. Marketing Res. 18(1981), 39.


A general model for production-transportation planning in steel supply chain

A. Sabzevari Zadeh1* and R. Sahraeian2

Industrial Engineering Department, Faculty of Engineering, Shahed University, Tehran, Iran

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------AbstractThis paper is focused on the tactical design of steel supply chain (SSC). A general mathematical model is proposed to integrate production and transportation planning in multi-commodity SSC. The main purpose is to prepare a countrywide production and distribution plan in an SSC with three layers consisting of iron ore mines as suppliers, steel companies as producers, and subsidiary steel companies as customers. An internal supply chain consisting of melting furnaces and casting lines in each producer is also taken into account. Demand is assumed to be deterministic and known at the beginning of planning horizon. Each mine supplies iron ore with specific chemical compounds (CCs) and each producer needs a given volume of iron ore with a pre-determined range of CCs which can be provided by mixing several kinds of iron ores. Finally, a real test case in SSC is described and solved using the proposed model. Sensitivity analysis supports the decision makers with noteworthy information about tactical decisions.

Keywords: Steel Supply Chain, Production, Transportation, Blending Problem, Linear Programming. -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

* Corresponding author:Tel: +98 (916) 6225761, Fax: +98 (611) 3362760 E-mail: [email protected]: Industrial Eng. Dept., Faculty of Eng., Shahed University, Tehran, 18155/159, Iran.1. MSc. Industrial Engineering2. Assistant Professor

1. IntroductionSupply chain is a network of facilities consisting of suppliers, producers, assembly lines and distribution centers. Material, information and financial flows interconnect them 1). It can be inferred that besides production and distribution tasks, a supply chain also consists of transportation, storage and retail activities 2).Decision making in supply chain management can be actuated in three distinct phases based on planning horizon: strategic, tactical, and operational 2). In strategic level, an organization plans the configuration of its supply chain for several future years. Strategic decisions include: determination of the tasks to be accomplished in the organization, and tasks to be outsourced, location and capacity of production and storage facilities, assignment of products to production and storage facilities, and transportation mode selection for shipping products. Planning horizon in tactical level is confined to some seasons or at most to one year. In this level, the organization decides about transportation and storage facilities which can be used regarding imposed restrictions in strategic framework. Finally, operational decisions concern flow shop or scheduling tasks for a special day or scarcely a week.This paper is focused on tactical decisions in steel

supply chain management. A mixed integer linear programming (MILP) model has been proposed to design and improve the production and transportation plan in a countrywide SSC. The proposed model determines the production levels in each producer, and transportation flows from each layer (or intra-layer) to the next one. The output of the model also consists of iron ore suppliers selection, production capacities assignment, and transportation planning for delivering products to the customers. So, the following questions would be replied through the proposed model:● How would the producers procure their required iron ore from the mines? ● How much of each product should be produced in each producer?● How much of each product should be transported from each producer to each customer?This paper is organized as follows. The problem of interest and its assumptions are described in detail in section 2. Section 3 presents the proposed MILP model. A real test case in SSC and its results are discussed in section 4. Finally, section 5 summarizes the paper and reveals some promising future research issues.

2. Problem definitionWe studied the dynamic multi-commodity production and transportation planning problem with deterministic demands and unlimited budget in SSC management. A supply chain network like the one depicted in Fig. 1, with three main layers comprising iron ore mines, steel producers and customers is considered. An internal supply chain with two intra-layers includes melting


12

Fig.1. Example of a SSC with an internal supply chain in each producer.

furnaces (i.e., blast furnaces or electric arc furnaces) and casting lines in each steel producer is also taken into account 3).The problem of interest consists of making decisions on production levels in producers' layer, efficient allocation of production capacities to the customers and product flows from each layer (or intra-layer) to the consequence one.Each mine supplies iron ore with a specific amount of CCs including Fe, Fe3O4, SiO2, Al2O3, S, P, CaO, and MgO. Moreover, each producer requires iron ore with a given range of CCs to be able to continue its robust process. In other words, each producer has a predefined standard to determine the admissible range of CCs of iron ore, empirically. It is important to note that the amount of each CC in iron ore affects process costs. As an example, sulfur (S) in any form is a temporary poison to reforming catalysts in direct reduction plants. Sulfur is present in iron ore and natural gas. Studies show when the amount of sulfur in iron ore increases, the reformer capacity will decrease. Therefore, the process costs increase by enhancing the amount of sulfur in iron ore. On the other hand, the iron ore price depends on its quality, so if a steel producer wants to supply iron ore with low sulfur, it has to pay much money. Admissible range of CCs in each producer is referred to the process technology and is specified by process experts empirically. Hence, each producer has to procure its required iron ore from several mines and mix them together to satisfy its CCs' range.The steel making process in each steel producer consists of three stages: smelting (i.e., blast furnace (BF) or electric arc furnace (EAF)), refining (i.e., ladle furnace (LF)) and continuous casting. In this paper, both smelting and refining stages are considered as one intra-layer in the steel producers. Moreover, there is an individual continuous casting line to produce each product. Due to flexibility in steel making processes, the production capacity of melting furnaces is assumed less than total capacity of the casting lines. For more detailed information about steel making process, interested readers could refer to Atighechian et al. 4) and Zanoni and Zavanella 5).

Steel products can be imported by customers and can be exported by producers. In other words, each customer is allowed to purchase steel products from an outside producer and each producer is allowed to sell some products to an outside customer. But, total volume of imported (exported) products by all customers (producers) at each time period has been limited. A lower bound on the ratio of production volume to the production capacity is defined as minimal acceptable utilization rate to guarantee the minimum level of benefit in each producer regarding its investment size.It is supposed that there are some steel producers in the SSC. Moreover, producers are crude steel companies, and customers are subsidiary steel companies. At the beginning of planning horizon, each producer has an initial capacity in smelting stage and casting lines. The planning horizon is divided into a set of consecutive and integer time periods which have equal lengths.

3. Problem formulationTo formulate the problem as an MILP model, the following notations are introduced in advance. Using these notations, a mathematical programming model can be formulated to solve the problem. It is also supposed that prior to model design; all relevant data were collected using e.g. appropriate forecasting methods and company-specific business analysis.

3.1. Notations

Sets:

I: set of iron ore minesJ: set of steel producersK: set of customersP: set of productsT: set of time periodsH: set of CCs of iron ore

Parameters:

t

iSC : maximal supply capacity of iron ore by mine i in time period t.

t

jPC : capacity of molten steel production in producer j in time period t.

t

pjCC : casting line capacity of product p in producer j in time period t.

jU : minimal acceptable percentage of utilization of producer j.

tI : maximal acceptable ratio of import to the total production volume in time period t.


13

tE : minimal acceptable ratio of export to the total production volume in time period t.

thisQ : percent amount of compound h in iron ore of mine i

in time period t.t

hjpQ : percent acceptable amount of compound h in required iron ore of producer j in period t.

tiYP :production yield) iron ore to steel products (for iron

ore of mine i in time period t.

jiCS : cost of purchasing and shipping per unit of iron ore from mine i to producer j.

pjCP : unit production cost of product p made by producer j.

pkjCT : cost of shipping per unit of product p from producer j to customer k.

tpkD : demand of customer k for product p in time period t.

pkiC : total cost for importing and shipping per unit of product p to customer k.

pjeC : net income for exporting per unit of product p by producer j.

−

+=

1

1

hI

if h indicate compound of Fe in iron ore,

otherwise.

Decision Variables:tjix : volume of iron ore purchased and shipped from mine i

to producer j in time period t.t

pjy : volume of product p produced by producer j in time period t.

tpkjz : volume of product p shipped from producer j to

customer k in time period t.t

pkimp : volume of product p imported by customer k in time period t.

tpjexp : volume of product p exported by producer j in time

period t.

3.2. Linear programming modelIn terms of the above notations, formulation of the problem is as follows.(M1) Minimize

tpkj

t j k ppkj

t k p

tpkpk

t j p

tpjpj

t j p

tpjpj

t i j

tjiji

zCT

impiC

eC

yCP

xCS

.

.

exp.

.

.

∑∑∑∑

∑∑∑

∑∑∑

∑∑∑

∑∑∑

+

+

−

+

(1)

Subject tot t tjkp kp kp

jz imp D+ ≥∑ , ,t k p∀ (2)

∑∑ ≥p

tpj

it

i

tji y

YPx ,t j∀ (3)

∑∑ ≥i

tji

thjh

tji

i

thih xpQIxsQI .... , ,t j h∀ (4)

t

ij

tji SCx ≤∑ ,t i∀ (5)

∑ ≥p

t

jjt

pj PCUy . ,t j∀ (6)

∑ ≤p

t

jt

pj PCy ,t j∀ (7)

t

pjt

pj CCy ≤ , ,t j p∀ (8)

∑+=k

tpkj

tpj

tpj zy exp , ,t j p∀ (9)

tt tkp jp

k p j pimp I y≤∑∑ ∑∑ t∀ (10)

exp tt tjp jp

j p j pE y≥∑∑ ∑∑ t∀ (11)

0exp,,,, ≥tpj

tpk

tpkj

tpj

tji impzyx , , , ,t i j k p∀ (12)

In the above formulation, objective function (1) minimizes the total costs including transportation and purchasing costs of iron ore from mines to producers, production costs, attained net income by export (with negative sign), import costs and transportation costs from producers to customers, respectively.Constraints (2) ensure that the demand of each customer is fulfilled. In order to provide the required molten steel proportional to customers' demands, constrains (3) assure that each producer receives enough volume of iron ore from all mines. Inequalities (4) ensure that each producer attains its required iron ore in conformity with its admissible range of CCs in each time period. Thus, these constrains enforce the blending problem into the proposed model 6,7).Constraints (5) impose the supply capacity on each iron ore mine. Constraints (6) ensure that output of each producer is more than a given minimum level. Constraints (7) state the maximum smelting stage capacity in each producer. Constraints (8) state the maximum casting line capacity of each product in each producer. Equations (9) impose the product flows conservation of producers while the demands of customers are gratified.Constraints (10) ((11)) state that the total volume of


14

imported (exported) products must not be more than (less than) a given maximum (minimum) level at each time period. Note that, these levels are legislated by government. Finally, constraints (12) enforce the non-negativity restrictions on the corresponding decision variables.

3.3. Model developmentThere are some actual issues in the real world SSCs which have not been considered in the formulation of the problem. For example, it is not possible to purchase iron ore in small batches, e.g. Chadormalo mine in Iran would contract to supply at least 300 thousand tons of iron ore per year. In order to enhance this matter to formulation, the following lemma can be used, efficiently.

Lemma: In order to confine the continuous variable x such that it takes either the value of zero or a value greater than or equal to K, new binary variable u and the following constraints set are introduced:

uMx .< (13)

uKx .≥ (14)

In which K is the minimum admissible batch size and M is a big number which should be at least greater than the sum of all supply capacities of iron ore mines.

Proof:a) If u=0, we will have 0<x according to constraint (13) and 0≥x according to constraint (14), which means x= 0, andb) If u=1, we will have Mx < according to constraint (13) and Kx ≥ according to constraint (14). Since M is greater than the biggest possible value for x, constraint Mx < is not an active constraint and the model can be solved regardless of this constraint. So, we will have Kx ≥ .□In order to exert this issue into the formulation, the three following constraints are appended to model M1.

tji

tji uMx .< , ,t i j∀ (15)

tji

ti

tji uKx .≥ , ,t i j∀ (16)

{ }0,1tiju ∈ , ,t i j∀ (17)

By applying the above mentioned realistic issue in the problem, the new MILP model (namely M2) is obtained.(M2) Minimize Objective function (1) Subject to Constraints (2) till (12), Constraints (15) till (17).A large portion of complexity of the above model is related to minimum admissible batch size constraints (i.e., constraints (15), (16) and (17)). So, when the number of iron ore mines or crude steel producers increases in the problem, the complexity of the

model will increase. It is important to mention that the proposed model is not very difficult to solve with standard solvers (e.g., LINGO, Cplex, Xpress, etc.) in small and medium test problems. Because, based on our recent investigation, the problem is not more general than any known NP-hard or NP-complete problems like location-allocation problem (LAP), traveling salesman problem (TSP) and etc. So, the proposed model does not belong to these classes. However, the complexity of the model increases by enhancing the number of binary variables (i.e., t

jiu ).

4. A case: Iran SSCTo assure the correct performance of constraints and objective function, the model is applied to a realistic scenario in Iran SSC where the customers' demands are deterministic. In the discussed SSC, domestic iron ore mines are Chadormalo, Golgohar, and Choghart; and exotic mines are Samarco, Kudermukh, Carajase, CVRD, Ferteco, and MBR. As mentioned in section 2, each of these mines supplies iron ore with specific CCs. By using CCs of each mine for some time periods, a regression model can be applied to forecast CCs at some time periods in the future.Steel producers' layer consists of Khouzestan Steel Co. (KSC), Mobarakeh Steel Co. (MSC), Esfahan Steel Co. (ESC) and an outside producer (OUT) to import products. Customers' layer comprises 1) MSC, 2) ESC, 3) Kavian Steel Co., 4) National Industrial Steel Group Co., 5) Ahwaz Pipe and Rolling Mills Co., 6) Khuzestan Oxin Steel Co., 7) Azerbayjan Steel Co., 8) Khorasan Steel Co., 9) Amir Kabir Khazar Steel Co., 10) Iran Alloy Steel Co. and 11) an outside customer to export products.It is assumed that there are three inside and one outside crude steel producers in the chain such that their capacities are expanded proportional to customers' demands. Materials flowing from the first layer to the second layer consist of iron ore with different CCs and from the second layer to the third layer which consist of three kinds of steel products namely; Billet, Bloom and Slab. The value of production yield from iron ore to steel products depends on total amount of Fe metal in iron ore and has to be less than 1.65. It means that the steel production process converts 1.65 tons iron ore to 1 ton steel products.The demand of the customers was estimated using moving average based on the available data from their demands in the last two time periods. Since steel is a functional and strategic commodity, consumer demand can be estimated with a high reliability 8).Import policy should be exerted such that the ratio of imported products to the total domestic demands at each time period becomes less than or equal to 5 percent. Instead, export policy should be exerted such that the ratio of exported products to the total domestic crude steel production becomes more than or equal to 10 percent at each time period.


15

Each steel producer has some melting furnaces (e.g., blast furnaces, electric arc furnaces or etc.) which feed some parallel casting lines. MSC is only able to produce slab, and ESC produces both billet and bloom. But, KSC and OUT can produce all three kinds of steel products. As noted earlier, due to flexibility in production, the capacity of melting furnaces is less than the total capacity of casting lines. For example, KSC has a capacity of 260 thousand tons per month in melting furnaces and its capacity in billet, bloom, and slab casting is 120, 120 and 100 thousand tons per month, respectively. Since this is the dominant scenario among steel producers, smelting stage has been considered as the bottleneck of steel production process.The main tactical objectives for modeling and analyzing this countrywide case are to plan for importing/exporting products, allocation of customers to the producers, iron ore suppliers selection for long term contracts and transportation planning.To solve the proposed model by LINGO11.0, the required data were gathered in advance and were recorded in an Excel sheet subsequently. The data were transferred from Excel sheet to LINGO environment by using OLE function. The Branch and Bound (B-and-B) solver was applied to solve the model and the Dept First strategy was chosen for node selection. After solving the model by LINGO11.0 some noteworthy results are obtained. Fig. 2 demonstrates production plan for three kinds of products at 5 time periods in the future. As can be seen in this figure, total production volume of billet by KSC and ESC at the first time period is 215 thousand tons. Since, total inside demands for billet is 190 thousand tons; 25 thousand tons (equivalent to 11.6 percent of domestic production) of this product should be exported by KSC to satisfy the constraint of minimum export volume. Instead, total inside demands for bloom is 205 thousand tons at the first time period, whereas total domestic steel production volume is 195 thousand tons. Thus, it is necessary to import 10 thousand tons (equivalent to 4.9 percent of total inside demands) to satisfy demands of this product and the constraint of maximum import volume. According to output of the model, National Industrial Steel Group Co. should be supplied a portion of its required bloom at the first time period by importing this product.

Fig.2. Planned production volume for three kinds of products.

Fig. 3 shows the assignment of consumers to the billet’s producers at the first time period. Based on this figure, total demands of inside consumers for billet are supplied by the domestic production. As an example, the required billet for ESC is supplied by itself, whereas National Industrial Steel Group Co. provides its required billet by purchasing 30 thousand tons from KSC. In general, the aforementioned results are obtained based on import/export policies, producers’ capacities and transportation costs between producers and customers.

Fig.3. Proposed plan for allocating customers to billet’s producers in Iran SSC.

5. Conclusion and future researchIn this paper, a general innovative model has been proposed for multi-period, multi-commodity production and transportation planning problem in SSC management. Our approach can be used both to design new networks and to improve existing networks. The main contributions of our work are to consider intra-layer product flows, blending problem and modeling a real case in steel industry. Moreover, determination of production levels and efficient allocation of customers to the producers are also taken into our consideration. In a practical application, Iran National Steel Co. can apply the proposed model to prepare production planning for steel producers and exert its import/export policies to reduce total costs in Iran SSC at a multi-period planning horizon.We have assumed that the admissible ranges of CCs in all producers have been defined prior to model design empirically. So, in order to design a more realistic model, a case can be considered in which these ranges are unavailable. The essential work in this case is investigation of the influence of CCs on process costs with statistical tools like regression models or design of experiments in advance; and subsequently, optimizing the total costs consist of steel production process costs and supply iron ore costs (e.g., purchase and transportation costs) by designing and solving a multi objective model.In this paper, it was also assumed that there are some steel producers whose capacities are expanded proportional to the deterministic demands. But, locating new steel producers has not been considered in the problem. So, exerting inventory and facility


16

location with stochastic demands in the problem is an interesting new area for research.

References[1] E.H. Sabri and B.M. Beamon: Omega, 28(2000), 581.[2] S. Chopra and P. Meindl: Supply chain management: strategy, planning, and operation, (3rd edition), New Jersey: Prentice Hall, (2007), 42.[3] M.T. Melo, S. Nickel and F. Saldanha da Gama: Eur. J. Oper. Res., 196(2009), 401.[4] A. Atighechian, M. Bijari and H. Tarkesh: Comput.

Oper. Res., 36(2009), 2450.[5] S. Zanoni and L. Zavanella: Int. J. Prod.Econ., 93(2005), 197.[6] B. Bilgen and I. Ozkarahan: Int. J. Prod.Econ., 107(2007), 555.[7] L. Chiun-Ming and Hanif D. Sherali, C.M.Liu and H.D. Sherali, Omega, 28(2000), 433.[8] M.L. Fisher: What is the right supply chain for your product, Harvard Business Review, March-April, (1997).

Influence of Aging Temperature on Mechanical Properties and Sound Velocity in Maraging Steel M350

P.Behjati1*, A. Najafizadeh2, H. Vahid dastjerdi3, M.Araghchi4, R.Mahdavi5

Materials Engineering Department, Isfahan University of Technology, Isfahan 84156-83111, Iran

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------AbstractIn the present work, the influence of aging temperature on mechanical properties and sound velocity of Maraging steel M350 was investigated. For this purpose, first, samples were solution annealed at 825◦C for 2 hours and then age hardened at 510◦C-600◦C for 3 hours. Hardness, tensile and impact tests were used for determining mechanical properties and longitudinal ultrasonic velocity was used for determining sound velocity of the samples. The obtained results indicated that hardness, strength and sound velocity of the samples decreases with increase of ageing temperature, whereas, toughness of the samples increases directly with ageing temperature. These results were attributed to the dissolution of Ni3(Mo,Ti) and Fe2Mo precipitates and also to the formation of reverted austenite which are promoted by the increase of ageing temperature. Optical microscopy of the samples revealed that, in this case, the morphology of reverted austenite is mainly of grain boundary and interlath type. Further, a linear correlation between the mechanical properties and sound velocity of the samples was found that can be used in industrial applications.

Keywords: Maraging steel, Ageing, Reverted austenite, Sound velocity.-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

* Corresponding author:Tel: +98 (311) 3915742, Fax: +98 (311) 3912752E-mail: [email protected]: Materials Engineering Department, IsfahanUniversity of Technology, Isfahan 84156-83111, Iran1. PhD. Student2. Professor 3. M.Sc. 4. M.Sc. 5. B.Sc.

1. IntroductionTraditionally, destructive tests such as tensile testing or metallographic inspection are used to control the quality of the steel products and to prove that the properties are the same as the design stage. Sample preparation of these techniques is done through cutting a representative sample from the material and performing tests on this sample. Despite the necessity of having precise knowledge about the actual material properties used in the structure, destruction of the material is impossible in most cases due to safety requirements. Also, high testing costs should be added to drawbacks of destructive tests. Therefore, determining the properties of materials with nondestructive techniques has attracted the researchers’ attention in recent years. Maraging steels with unique combination of properties such as ultrahigh strength, high toughness, good formability and weldability have received attention of many strategic industries 1). These steels receive their strength and toughness from martensitic transformation followed by age hardening. Since the martensitic transformation only involves an austenitic

to martensitic transformation of Fe-Ni, and does not involve carbon to any considerable extent, the martensite which is formed is relatively ductile 2). The level of strength attained during age hardening depends on the time and temperature of ageing process. Results of earlier studies show that the peak hardness and strength of these alloys is achieved by the formation of metastable Ni3Mo, Ni3Ti and Fe2Mo precipitates 3-7). Prolonged ageing or overageing at higher temperatures results in the formation of reverted austenite 8). Few investigations have been performed to determine the properties of maraging steels using nondestructive techniques 12,13). The main purpose of this work is to find out the influence of aging temperature on mechanical properties and sound velocity of Maraging steel M350.

2. Experimental procedureIn this work, raw material of M350 steel was provided in the form of extruded bar with a diameter of about 60 mm. Composition of the alloy is given in Table 1. Raw material was initially solution annealed at 825 °C for 2 hr in a vacuum furnace. Then, samples for tension and impact tests were prepared from the annealed material according to the ASTM E23 and ASTM E8 standards, respectively. After that, ageing treatment was carried out on samples at temperature range of 510 to 600 °C for 3 hr in the same vacuum furnace. Three tensile samples were tested for each ageing condition at a strain rate of about 0.01 s-1. On each sample, 5 hardness values were taken and the average value was reported. An optical microscope (OM)



18

was used to study the microstructure of samples. Specimens for OM study were first mechanically polished and then etched using an aqueous solution consisting of 30 ml H2O, 5 gr CuCl2, 40 ml HCl and 25 ml ethanol. Pulse-echo technique was used to measure the ultrasonic velocity within the samples. Longitudinal ultrasonic wave was produced using a 20 MHz TR normal probe. 3. Results and discussionFig. 1 shows the variation of mechanical properties with ageing temperature. As shown in Figs. 1a and 1b, the maximum hardness and strength of the alloy (2361 MPa and 730 HV, respectively) and also the minimum toughness (9j) is achieved after 3 hr of ageing at 510°C. Also, results of TEM studies 6,13) show that ageing of this alloy at 510°C leads to the formation of Ni3(Ti,Mo) and Fe2Mo precipitates. These precipitates have an ellipsoidal morphology and increase strength of the alloy. Increase in temperature results in the dissolution of these strengthening precipitates in the matrix which locally enhances the concentration of Ni element, producing preferential sites for nucleation of austenite phase. In Fig. 1,it can be seen that hardness and strength values decrease with an increase in ageing temperature, whereas, toughness values decrease.Viswanthan et al14). suggested that this trend be attributed to enhanced formation of retained austenite with an increase in ageing temperature. Depending on the temperature, time and rate of heating different morphologies of austenite (such as grain boundary, interlath and Widmanstätten) may be produced in the martensitic matrix 14-16). In the present work, microstructural study of the samples showed mainly grain boundary and interlath morphologies of austenite. Fig. 2 shows typically microstructure of a sample aged at 600°C. Small and large arrows represent grain boundary and interlath morphologies of austenite, respectively. Fig. 3 shows the variation of sound velocity with ageing temperature. It is observed that increase in temperature brings about constant decrease in sound velocity. Rajkumar et al. and Yeheskel11,12) investigated the effect of age hardening on sound velocity in maraging steel 250. They found that with the onsetof retained austenite formation, the longitudinal sound velocity decreases severely. They attributed this behavior to the increase in Ni concentration in the matrix that reduces elastic modulus. Regarding the relationship between elastic modulus (C), sound velocity (v) and density (ρ) of metals (C= ρv2), elasticmodulus has a considerable effect on sound velocity.

(a)

(b)

Fig. 1.Variation of (a) hardness and strength values,and (b) toughness values of the samples with ageing temperature.

Fig. 2.Optical microstructure of the sample aged at 600°C.

Table 1. Chemical composition of M350 steel used in this work.

Element Ni Co Mo Ti Al Mn Si C Fe%Wt 17.88 12.03 3.93 1.54 0.17 0.03 0.05 0.004 .Bal


19

Fig. 3.Variation of the sound velocity with ageing temperature.

One of the important findings of this research is the linear relationship between mechanical characteristics and sound velocity of maraging steel 350. Fig. 4a shows the linear relationship between sound velocity, hardness and strength with a positive slope. Fig. 4b shows the linear relationship between the sound velocity and toughness of the alloy which conversely has a negative slope. This difference in the relationships is related to the properties of austenite phase. This phase has an FCC crystal structure, high ductility and low hardness. On the other hand, formation of retained austenite enhances the Ni concentration in the matrix, decreasing elastic modulus and sound velocity of the alloy. These linear relationships can be used for determining the characteristics of heat treated parts or even controlling of heat treatment processes nondestructively, and therefore are important from an industrial viewpoint.

4. Conclusion In this research the influence of ageing temperature on mechanical characteristics, ultrasonic velocity and also their relationships were investigated in maraging steel M350. It was found that with increase in ageing temperature strength, hardness and sound velocity of the samples decrease, however, toughness of the samples increases. These trends were related to the enhanced formation of retained austenite with increases in ageing temperature. This phase has an FCC crystal structure, high ductility and low hardness. On the other hand, formation of retained austenite enhances the Ni concentration in the matrix, decreasing elastic modulus and sound velocity of the alloy. Further, a linear relationship between mechanical properties and sound velocity of the samples was found, which is important from an industrial viewpoint.

(a)

(b)

Fig. 4 .Relationship between sound velocity and (a) hardness and strength, and (b) toughness of the alloy.

Reference:[1] R.F. Decker, J.T. Eash, A.J. Goldman: Transactions of ASM, 1962, 55,1-19.[2] W.F. Smith, “Structures and properties of engineering alloys”, 1981, New York, McGraw-Hill, 157-9.[3] X. Li, Z. Yin: Mater. Lett., 23(1995), 269.[4] X. Li, Z. Yin: Mater. Lett., 24(1995), 235.[5] F. Habiby, T.N. Siddiqui, H. Hussain, A. UL Haq, A.Q. Khan, Mater. Sci., 31(1996), 305.[6] R. Tewari, S. Mazumder, I.S. Batra, G.K. Dey, S. Banerjee, Acta Mater.,48(2000), 1187.[7] E.V. Peterloma, A. Shekhter, M.K. Miller, S.P. Ringer, Acta Mater.,52(2004), 5589.[8] N. Atsmon, A. Rosen, Metallography, 14(1981), 163.[9] K.V. Rajkumar, B.P.C. Rao, B. Sasi, Anish Kumar, T. Jayakumar, Baldev Raj, K.K. Ray, Mater. Sci. Eng. A,464(2007), 233.


20

[10] S.H. Khan, A. Nusair Khan, F. Ali, M.A. Iqbal, H.K. Shukaib, J. Alloys. Compd., 474(2009), 254.[11] K.V. Rajkumar, Anish Kumar, T. Jayakumar, Baldev Raj, and K.K. Ray, Metal. Mater. Trans. A, 38(2007), 236.[12] O. Yeheskel, Metal. Mater. Trans. A, 38(2007), 236.[13] U.K. Viswanthan, G. K. Dey, M. K. Asundi,

Metal. Trans. A, 24(1993), 2429.[14] U.K. Viswanthan, G. K. Dey, V. Sethumadhavan, Mater. Sci. Eng. A, 398(2005), 367.[15] Rajeev Kapoor, I.S. Batra, Mater. Sci. Eng. A, 371(2004), 324.[16] M. Farooque, E. Ayub, A. UL Haq, A.Q. Khan, J. Mater. Sci., 33(1998), 2927.

Characterization and Catalytic Behaviour of Nanostructured Iron Oxide Powder from Waste Pickle Liquor of Steel Industry

A. Hosseini1*, M. Alizadeh2

Materials and Energy Research Center, P.O. Box 31787-316, Karaj, Iran

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- AbstractNanostructured iron oxide powder that has been recovered from waste pickling liquor unit of steel industry was studied for oxidation catalytic applications. In this research, the powder was characterized by X-ray diffraction (XRD) and transition electron microscopy (TEM) for determination of phase structure, morphology and particle size. Furthermore, the specific surface area of the powder was achieved by BET method. Results show that the powder consists of α-Fe2O3 particles with wide range distribution particle sizes between 50-300 nm. Specific surface area of the powder is about 4.33m2g-1. Moreover, catalytic behaviour of the iron oxide powder was investigated. A quartz tubular furnace was applied as a reactor for measuring catalytic activity. Conversion test results show that CO to CO2 conversion reaction was initiated at about 300°C (light-off temperature) and was completed at about 600°C. Also,the CO to CO2 conversion ratio was almost constant during long-term stability test.

Keywords: Iron oxide powder, characterization, Oxidation catalyst, CO removal-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

* Corresponding author:Tel: +98 (261) 6210700, Fax: +98 (261) 6200777E-mail: [email protected]: Materials and Energy Research Center, P.O. Box 31787-316, Karaj, Iran1. PhD Student2. Assistant professor

1. IntroductionExhaust gas that exits from the automobile engine and some furnace stack contains poisonous components, such as carbon monoxide and hydrocarbon volatile gasses. Oxidation catalytic materials could convert these harmful components to inert gases e.g. carbon dioxide and water. Nowadays Pt, Pd, and other platinum group metals (PGM), as an active catalytic sites are used for the removal of poisonous gasses. However, due to the rising cost of PGM, many researchers have been searching for alternative materials as the active catalytic phase. Use of nanostructure and nanoparticle transient metallic oxides for catalytic applications was studied in the last decade 1-5). These compounds are promising materials as an oxidation catalyst due to low cost, stability in high temperature and high sintering temperature. Hematite (Fe2O3) is one of transient metal oxides that is well-known with an unusual magnetic behavior. It is used for different applications such as pigment for paints, biomedical application, magnetic component fabrication, removal of heavy metals from aqueous solutions, and catalysts for petrochemical industries 6-10). Fe2O3 nano-powder is acknowledged as a suitable material for oxidation catalytic systems. Various methods such as sol-gel, chemical precipitation, and high-energy ball milling have been used to produce

nanostructure and nano-particles iron oxide 9-12). Li et al. have investigated catalytic behavior of Fe2O3 with 3 nm particle sizes (NANOCAT) for the removal CO from pollution gases. Experimental results show CO to CO2 conversion reaction was completed at 350 °C. Also, the conversion reaction kinetics was first-order with respect to CO 3). Similar results were reported for this powder by Kwon et al. 2). Halim et al. have studied the catalytic behavior of different sizes of nano-crystalline and nano-particles of iron oxide powders that are prepared by sol-gel method 4-5). In this research, iron oxide powder that has been achieved from waste pickling liquor unit of steel industry was examined for oxidation catalytic behavior. For this purpose, XRD, TEM and BET methods were used to characterize this material.

2. Materials and methodThe powder structure was characterized by X-ray Diffraction (XRD, Philips Expert, Germany) with Cu Kα (λ = 1.5406Å) radiation in 2θ range from 5 to 80°. Transition Electron Microscopy (TEM, Philips CM200 FEG, Germany) was used for the investigation of morphology and size of the powder. The specific surface area of the catalyst powder was obtained by Brunauer- Emmett-Teller method (BET, Bel-Belsorp mini, Japan). For measuring catalytic activity, a quartz tubular furnace was used. Glassy wool was impregnated by catalyst materials and then, was inserted in the furnace. An amount of 500mg Fe2O3 powder was used as an oxidation catalyst material. Furnace temperature increases gradually by rate of 3°C/s. A mixed gas that consists of 1%O2-0.5%CO-98.5%N2 was passed



22

through the furnace, and CO concentration in the outlet flow gas was detected by gas analyzer instrument (Testo 327-1, Germany). CO to CO2 Conversion ratio was calculated as follow:

CO to CO2 Conversion Ratio int

int

%%%

OCOCOC ext−

= (1)

Also, long-period conversion test was performed to investigate the conversion reaction stability. The stability test was carried out for several temperatures as long as 450 minutes.

3. Results and discussion3.1. Powder charactrizationDuring hot rolling process, iron oxide layer forms on surface of steel slab. This layer must be removed by risining process after hot rooling. The acidic solution obtained in this stage is called waste pickle liqour (WPL) and is transfered to acid recycle unit. Then, WPL is sprayed into the furnace (spray roasting process) to recover acid solution. In this stage, iron chloride solution is converted to Fe2O3 powder in the presence of water vapor and oxygen at high temperatures according to the following reactions 13):

2FeCl2 + 2H2O + ½O2 → Fe2O3 + 4HCl (2)

2FeCl3 + 3H2O → Fe2O3 + 6HCl (3)

Subsequently, submicron red-brown iron oxide powder was formed. Fig. 1 shows the micrograph of the iron oxide powder. This Iron oxide powder was commercially used as a yellowocher pigment dye in paint industry 6).

Fig. 1. Micrograph of the iron powder used in this research.

Fig. 2 shows X-ray diffraction pattern of the iron oxide particles. All of the sharp and narrow peaks in this figure are related to the α-Fe2O3 phase according to Joint Committee on Powder Diffraction Files (JCPDF card No. 24-0072). This result indicates the purity and highly crystalline of the Fe2O3 particles.

Crystallite size has been calculated by using Scherer equation:

(4)

where D is the mean crystallite size of the powder, λ is the wavelength of Cu kα equal to 1.54056Å, β is the full width at half-maximum (FWHM) intensity of the (104) peak in radian that is measured by X‘Pert HighScore software, θ is Bragg’s diffraction angle and K is a constant usually equal to 0.9. According to this equation, averrage crystallite size is about 50 nm.

Fig. 2. X-ray diffraction pattern of iron oxide nano-particles.

Micrograph of nanoparticles iron oxide obtained by transition electron microscopy is shown in Fig. 3. It is clear that the shape of the particles is spherical and their sizes vary from 50 to 300 nm. Because of conversion of CO to CO2 was done on the surface of the powder, specifiec surface area is an important parameter in oxidation catalytic materials. BET test results demonstrated that the specific surface area of the powder is 4.33 m2/g. Therfore, these particles can be used as oxidation catalyst for CO conversion to CO2 due to relatively high specific surface area.

Fig. 3. Transmission electron microscopic images of the iron oxide nano-powder.

3.2. Oxidation catalytic activityCO to CO2 conversion ratio versus temperature of

θβλ

=cosKD


23

iron oxide powder for 1%O2-0.5%CO-98.5%N2 gas mixture is illustrated in Fig. 4. It is obvious that catalytic activity increases with process temperature. As shown in this figure, a temperature of about 300°C is required for the initiation of oxidation reaction and no conversion reaction is detected at less than this temperature. Therefore, light-off temperature of conversion reaction is determined about 300°C. As CO to CO2 conversion reaction was completed at about 600°C, Fe2O3 nano-powder can catalyze oxidation of 100% carbon monoxide in the inlet gas mixture to carbon dioxide.

Fig. 4. Behaviour of CO to CO2 conversion at different temperature (500mg Fe2O3).

Overall reaction of CO to CO2 oxidation is:

2221 OCOOC →+ (5)

This reaction consists of two steps as follows 3):

21 OCOMOCOM xx +→+ − (6)

xx OMOOM →+− 21 2

1 (7)

As the above reactions show, CO molecules partially reduce metal oxide, and carbon dioxide molecule forms according to (6). In the second step, reduced metal oxide repeatedly absorbs oxygen from gas stream. Finally, catalytic material remains without change at the end of the process 3). It is clear that the presence of oxygen in passed gas stream is essential, so this is called oxidation catalyst system.A series of oxidation catalytic tests was performed in various CO concentrations of inlet gas mixtures. Oxygen concentration in inlet gas flow was considered constant in all tests. The effect of CO concentration on conversion ratio at 500°C was plotted in Fig. 5. As shown in this figure, by increasing CO concentration, conversion ratio linearly decreses. So, it indicates that

CO to CO2 conversion reaction is a first-order reaction with respect to CO concentration. Similar results were reported by other researchers 3).

Fig. 5. The effect of CO concentration on CO to CO2 conversion ratio at 500°C.

One of the main specifications of catalyst materials is their behavior in long-term service. In order to investigate this parameter, CO to CO2 conversion ratio was measured at 600°C for long time (450 min). Fig. 6 indicates that CO to CO2 conversion ratio is almost constant and it even gradually increases from 0.97 to 0.98 after 450 minutes.

Fig. 6. Stability of CO to CO2 conversion reaction at 600°C for 450 minutes.

4. ConclusionsIron oxide powder that has been precipitated from waste pickling liquor unit of steel industry is a suitable material for oxidation catalytic applications. Some aspects such as stability in high temperature, relatively high specific surface area, low cost, rapid fabrication procedure and ease of availability are the advantages of these powders. CO to CO2 conversion reaction is completed about 600°C for low concentration of CO in the gas stream. Also, the conversion reaction is stable for long-term service.


24

References[1] M. Twigg: Appl. Catal. B-Environ., 70(2007), 2.[2] S.C. Kwon, M. Fan, T.D. Wheelock and B. Saha: Sep. Purif. Technol., 58(2007), 40.[3] P. Li, D.E. Miser, S. Rabiei, R.T. Yadav and M.R. Hajaligol: Appl. Catal. B-Environ., 43(2003), 151.[4] K.S. Abdel Halim, M.H. Khedr, M.I. Nasr and A.M. El-Mansy: Mater. Res. Bull., 42(2007), 731.[5] M.H. Khedr, K.S. Abdel Halimb, M.I. Nasr and A.M. El-Mansy: Mater. Sci. Eng. A, 430(2006), 40.[6] M. J. Potter: Iron oxide pigments, U.S. Geological Survey Minerals Yearbook, (1999), 42.1.[7] C. Corot, P. Robert, J. Idée and M. Port: Adv. Drug

Delive. Rev., 58(2006) 1471.[8] F. Kenfack and H. Langbein: Cryst. Res. Technol. 41(2006), 748.[9] T. Osaka, T. Matsunaga, T. Nakanishi, A. Arakaki, Daisuke Niwa and H. Iida: Anal. Bioanal. Chem. 384(2006), 593.[10] K. Gupta and U. C. Ghosh: J. Hazard. Mater. 161(2009), 884.[11] L.Wang and J. JiangL: Physica B, 390(2007), 23.[12] A.S. Lileev, Yu.D. Yagodkin, E.N. Grishina, E.S. Khanenya, V.S. Nefedov and O.I. Popov: J. Magn. Magn. Mater., 290–291(2005), 1217.[13] W.F. Kladnig: Int. J. Mater. Pro. Tech., 21(2004), 555.

The effective parameters on thermal recovery and reduction of iron oxides in EAF slag

A.M.Taheri1*, A. Saidi2, A.A. Nourbakhsh3

Department of Materials Engineering, Islamic Azad University of Najafabad, Iran

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- AbstractSteelmaking slag of Mobarakeh Steel Company (MSC) has a high iron content which makes it unsuitable for conventional slag recovery; as in cement industries. On the other hand, EAF slag in this company is almost free of zinc because of using more than 90 percent DRI in the charge. Therefore, removal or recovery of iron from the slag can be a suitable solution to slag recovery in this company. In this research work, cold bonded briquette was made from mixtures of milled slag and coke breeze. The effect of binder content on the strength of the briquettes was studied. The reduction of iron oxides was also investigated by heating up the briquettes in a muffle furnace. The results indicated that using 4wt% sodium silicate as binder leads to reasonable compression strength of about 1500kg although the strength could be increased to as high as 2000kg by using 6wt% binder. Up to 40% reduction, the reducing rate was almost high; then it decreased and reached a minimum value at 70% reduction. Using extra carbon than the reaction stoichiometry didn’t have significant effect on the reduction. Temperatures above 1000°C up to 1200°C had also a small positive effect on the reduction rate. When reduction was performed at 1000°C, for 180min, semi-reduced material (70% reduced) could be obtained, which can be used (cold or hot) in BOF or other steelmaking processes.

Keywords: EAF slag, Briquette, Recovery, Reduction. -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

* Corresponding author:Tel: +98 (311) 3915713, Fax: +98 (311) 3912752E-mail: [email protected]: Dep. of Materials Engineering - Islamic AzadUniversity, Najafabad Branch. 1. M.Sc.2. Professor3. Associate Professor


1. IntroductionAn integrated steel plant generates approximately 420 kg of solid waste per ton of steel, consisting mainly of slag, dust and sludge. Slag is actually a secondary resource of metals, rather than an end-waste which is utilized as a resource material in many areas. The conventional method for disposal of slag is dumping 1). Thanks to the intensive research work during the last 30 years, about 65% of the steel slag is used in qualified fields of application, but the remaining 35% is still dumped 2). However, it is possible to use the dumped slag in different fields of application such as cement production, road construction and civil engineering, fertilizer production 3,4), landfill daily cover, soil reclamation and water refinement and production of metallic iron and iron concentrate 1) (Fig. 1).Steelmaking slag comprises about 10–15 wt % of the steel output 1). It means that each year more than 100,000,000 tons of steel slag is produced in the world. Chemical compositions of various steelmaking slags are summarized in Table 1.Steelmaking slag typically contains 20–25% iron.

which is normally in the form of iron oxide such as magnetite (Fe3O4) or other complex oxides with CaO or SiO2 2). Iron can be separated from slag by applying mineral processing technology and subsequently recycled as feed materials for the blast furnace. Due to high basic oxides (Mgo, CaO) in the slag, it can be used as a part of flux used in iron and steelmaking processes 1,2).The magnetic separation method can be used for separating metallic iron and iron minerals from steel slag 5). As mentioned above, steel slag is usually subjected to metal recovery prior to its application outside the iron and steel making process. The methods for slag processing are different, depending on the cooling method, chemical and mineralogical composition of steel slag, and its application. In general, steel slag processing includes crushing, grinding, screening and magnetic separation, and sometimes removal of phosphorous 2).In order to liberate metallic iron and iron-bearing minerals from steel slag, size reduction is necessary. By autogenous grinding, magnetic separation and screening, 8% metallic iron, 4 –7% iron concentrate and 85% ironless slag can be obtained. The metals in iron and steel slag are tightly bound to the slag matrix and not readily leached, and there are few environmental concerns regarding their application.However, by applying mineral processing technologies such as crushing, grinding, classification and magnetic


26

separation, it is possible to produce iron concentrate (Fe>55%) from steel slag. Carbothermic reduction of slag and sludge to produce semi reduced iron has also been reported 6).

2. Experimental procedure EAF slag from Mobarakeh Steel Company (MSC) and fine coke were used in this experiment as raw materials. Analysis of the slag (Table 2) showed that it contains 27.7%FeO and 7.3% Fe2O3. The slag was crushed, ground and screened to 325 meshes. Grounded slag was mixed with fine coke and sodium silicate as binder, and pressed into briquettes. Briquette making was carried out at 150 kg/cm2 pressure using a cylindrical die.Cold-bonded Briquettes were reduced under isothermal conditions in a muffle electric furnace. Reduction was carried out at different temperatures (1000-1200°C) with time intervals of 20, 60, 120 and 180 minutes. Also, the effect of extra carbon than reaction

stoichiometry (reaction 1, 2) was considered and briquettes with 0, 30 and 100 wt% excess carbon (fine coke) were made.Chemical analysis was carried

out of reduced briquettes to determine un-reduced FeO content. The weight changes of the samples (Δm) was also used to calculate degree of reduction according to equation (1):

R = 16 Δm / 28mO (1)Where:R= degree of reduction Δm= weight loss during reductionmO= weight of the oxygen in iron oxide of the sample (could be worked out from chemical analysis of slag). The Arrhenius equation (K = A e-E/RT) was used for the activation energy calculation; where, K = Rate Constant, A= Arrhenius Constant, E= Activation Energy, R= Gas Constant, T= Temperature.

Table 1. Chemical composition of steel slag (wt %) 2).

Fe total CaO MnO MgO SiO2 P2O5 Al2O3Slag Country

20–14 55–45 <5 <3 18–12 <2 <3BOF slag low MgO ontent Europe 20–15 50–42 <5 8–5 15–12 <2 <3BOF slag high MgO content

28–18 40–30 <6 8–4 17–12 <1.5 7–4EAF slag low MgO content 29–20 35–25 <6 15–8 15–10 <1.5 7–4EAF slag high MgO content 17.5 44.3 5.3 6.4 13.8 NR 1.5BOF slag (converter) Japan 15.2 38.0 6.0 6.0 19.0 NR 7.0EAF slag

(electric furnace oxidation) 27–17 48–34 6–1.5–2.5

10 15–9 0.9 2.8–0.9BOF slag China

30–15 50–40 10–5 10–5 15–10 3–1 2BOF slag USANR=no report.

Table 2. Chemical composition of EAF slag of MSC (wt.%).FeO Fe2O3 CaO SiO2 MgO Al2O3 MnO P2O5

27.7 7.3 29.2 19.1 10 3.9 2.2 0.6

Fig. 1. Utilization of steel slag 1).


27

3. Results and discussionFig. 2 shows the change of briquette strength with binder dosage. As seen, compression strength of the briquette increases with increasing the binders content, and it reaches a constant value of 2000 kg at 8 wt% binder. Since compression strength of 1500 kg is adequate for most industrial reduction processes, 4 wt% blinder were used to prepare samples for reduction tests.The process parameters of the reduction experiments were reaction temperature, time and excess carbon percent than stoichiometry value (EC%). Reduction experiments of the composite pellets were carried out at 1000, 1100 and 1200°C with different fine coke consumptions.

Fig. 2. Effect of binder dosage on the briquette compression strength.

Figs. 3 and 4 show the effect of temperature, time and EC% on the FeO % and degree of reduction, respectively. As seen in Fig. 4, the amount of reduction increases, for any EC ratio as the temperature increases. Also, longer reduction time is required at lower temperatures in order to reach the same reduction degree. The Effect of time on the reduction degree is higher than that of temperature. For all conditions, in order to reach a high degree of reduction (above 50%), the reduction time should be higher than 120 min.Other researchers reported that reduction of iron oxides occurs either by carbon or by carbon monoxide, where indirect reduction with CO requires higher temperature; since the carbon gasification reaction is highly endothermic 8,9). However, the results indicate that temperature in the range of 1000-1200 °C does not have a great effect on the reduction rate. Using excess carbon also does not have a significant effect on reduction. As can be seen in Fig. 5, the rate of reduction is relatively high at the first 40% of reduction. In fact, the first 40% of reduction takes place at 60 min while the next 20% reduction requires 120 min time. This situation is usually the case when reduction proceeds with diffusion.As was noted, reduction experiments were carried out at 30 and 100% EC. Fig. 5 clearly indicates that increasing this ratio causes a little increase in the degree of reduction.

Fig. 3. FeO% of cold bonded briquettes containing different EC% reduced at time and temperature.

Fig. 4. Degree of reduction of cold bonded briquettes containing different EC%.

Fig. 5. Effect of EC % on the degree of reduction with increasing time at 1000°C.

Kinetics of DR process:Reaction kinetics of iron ore reduction deals with the rate at which iron oxide is converted to metallic iron by removal of oxygen. The rate at which the ore is reduced influences the production rate, which ultimately determines the economic, feasibility, and competitiveness of the process technology involved. The reduction of iron oxide to metallic iron proceeds through various kinetic steps one of which is the slowest step, which controls the overall reaction rate .Different mathematical model equations are proposed to represent different rate controlling steps, which are given below:


28

The results of the isothermal reduction of slag for different time, temperature and EC% are shown in Fig. 4. In order to ascertain the appropriate kinetic equation, these results were checked against standard reduced time plots. Reduced time plots of experimental data are then superimposed on them in order to determine which theoretical plot fits the experimental data. Fig. 6 shows the theoretical plots for five kinetic laws and experimental points from the reduction data presented in Fig. 4 It is found that the following model would change linearly with time,

1- 2/3R - (1 - R) 2/3 = kt (2)This model shows that the reduction process was diffusion control. Fig. 7 shows the Arrhenius plot. Slope of the graph gives the value of activation energy which was calculated from Arrehenius Equation: K=Ae-E/RT.The activation energy of the process was calculated as 96.52 kJ/mole. In a number of other research involving the iron oxide-carbon system, kinetic data obtained were found to fit the Ginstling-Brounshtein model (1-2/3R - (1 - R) 2/3 = kt ) 7,8). The activation energy values they calculated seem to be lower than those found in this study. Utilization of slag in present work in comparison to iron oxides with higher purity is a reason for this.

Fig. 6. Experimental data compared with five different kinetic mechanisms.

Fig. 7. Arrhenius plot ln (K) vs 1/T×103.

4. Conclusion● Cold bonded briquette can be produced from the crushed steelmaking slag, using sodium silicate as a binder. With addition of 4 wt% binder, compression strength of about 1500 kg can be obtained. The strength reaches a maximum of about 2000 kg when 8 wt% (or more) binder is used.● Excess carbon above the stoichewmetric value doesn’t have a significant effect on the rate of reduction at the experimental conditions.● Temperature in the range of 1000 – 12000 °C has a little effect on the rate of reduction, being higher at higher temperatures.● The reduction process could be divided in to two stages. The reduction rate at the first stage (up to 40 % reduction) is relatively higher than that in the second stage. However, at 12000 °C a reduction degree of 70 % can be achieved after 180 min.

AcknowledgementsAuthors are grateful to the Steel Research Center of Isfahan University of Technology and Islamic Azad University of Najafabad for their financial support.

References[1] H. Shen, E. Forsberg: Waste. Manage. 23 (2003), 933.[2] S.K. Kawatra, S.J. Ripke: Int. J. Miner. Process. 65 (2002), 165.[3] M. Maslehuddin, A.M. Sharif, M. Shameem: Constr. Build. Mater. 17 (2003), 105.[4] A. Monshi, M.K. Asgarani: Cement. Concrete. Res. 29 (1999), 1373.[5] H. Alanyali, M. C. Yılmaz, S. Karago¨z: Waste Manage. 26 (2006), 1133.[6] H. K. Chen Lin, SH. Liu: Scand J Metallurgy: 21 (1992), 218.[7] M. Rahman, R. Haque and M.M. Haque: Ironmak. & Steelmak: 22 (1995), 166.[8] S. Mookherjee, H.S. Ray and A. Mukherjee: Ironmak. Steelmak: 13(1986), 229.[9] T. Sharma: Int. J. Miner. Process. 39 (1993), 299.

Equation Controlling step1- (1-R)1/3 = kt Chemically controlled- ln (1-R) = kt Chemically controlled[1 - (1-R) 1/3]2 = kt Diffusion Controlled1- 2/3R - (1 - R) 2/3 = kt Diffusion Controlledk’[1-2/3R - (1-R) 2/3]+D/r0[1-(1-R) 1/3]= kt Mixed Controlled

Quantitative measuring of pearlite in carbon steels using electromagnetic sensor

S. Kahrobaee1*, M. Kashefi 2, M.H. Nateq3

Department of Materials Science and Metallurgical Engineering, Engineering Faculty, Ferdowsi University of Mashhad, 9177948944-1111, Iran

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- AbstractNon-destructive Eddy current (EC) technique has long been used to detect discontinuities in materials. Recently, its application has been extended to characterize materials microstructure. In order to identify different microstructures, four plain carbon steel bars with different chemical compositions (AISI 1015, 1035, 1045 and 1080) were used in annealed condition. The pearlite percentage, carbon content and estimated hardness were determined according to responses of the samples to eddy current. They include primary and secondary voltages and normalized impedance. These data were compared with those obtained from conventional metallographic method and hardness measurements. The results show the high precision of the non-destructive eddy current method in determining the pearlite percentage, hardness and carbon content of mild carbon steels.

Keywords: Eddy current method, Materials characterization, Pearlite percentage, Hardness-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

* Corresponding author:Tel: +989155160084 Fax: +985118763305E-mail: [email protected]: Dept. of Materials Science and Metallurgical Engineering,, Engineering Faculty, Ferdowsi University of Mashhad, Mashhad, Iran.1. phd student2. Assistant Professor 3. M.Sc. student

1. IntroductionNowadays, application of non destructive methods is not limited to detect defects and cracks. Considering the advantages of these methods in quality control, in recent years several researches have been focused on non-destructive determination of the mechanical and physical properties of materials as a substitution for destructive methods. This new application for traditional eddy current techniques results in saving time and energy as well as providing 100% quality control in mass production line1). Among different methods, eddy current technique has individual advantages. Proper sensitivity to chemical composition, microstructure, mechanical properties and residual stress make it a reliable alternative to conventional destructive methods2,3).Recently, Konoplyuk managed to establish an appropriate relationship between the hardness of ductile cast iron and the primary and secondary voltages of eddy current signals4). Uchimoto and Check5,6), found the same relationship for gray cast iron, and they managed to determine mechanical properties of cast ductile iron such as elongation and tensile strength using nondestructive eddy current method. Besides, decarburizing depth was also studied using harmonic analysis in martensite base

microstructure of steel parts by Mercier et al7). Indeed using this nondestructive method, they have shown decarburizing depth can be measured after calibration of the Eddy Current system. Furthermore, on the basis of difference in magnetic properties of decarburized zone and the core of the mild carbon steel parts, the thickness of decarburized layer has been estimated using multifrequency electromagnetic sensor8). More recently, Rumiche et al. have investigated the effect of microstructure on magnetic behavior of carbon steels by electromagnetic sensors9), and the effect of grain size on magnetic properties was investigated and proved by other researchers10-12).The potential to determine measurable microstructure characteristics of steel parts has not been explored nondestructively. Therefore, the goal of the present study is to determine pearlite percentage, carbon content, hardness, and nondestructively according to magnetic responses of plain carbon steels to eddy current.

2. ExperimentalFor the purpose of determining the pearlite percentage in steel, four sample rods with 22mm diameter and 150mm length were prepared from four different kinds of steels (AISI 1015, AISI 1035, AISI 1045 and AISI 1080). Chemical compositions of steels are presented in Table 1. All samples were austenitized in 900°C for 30 minutes. Subsequently, all samples were cooled to ambient temperature resulting in equilibrium microstructures of pearlite and ferrite.Phase fraction percentage and the hardness of the samples were measured by metallographic method (using Microstructure Image Processing (MIP)


research note

software) and hardness destructive measurement method to compare with the obtained nondestructive values.Finally, Eddy Current tests were performed on the cylindrical samples at different frequencies. A schematic diagram of the used Eddy Current system is

Table 1. Chemical composition of studied steels

%P%Mn%Si%CSteel0.030.530.260.13AISI 10150.020.550.200.34AISI 10350.0130.570.300.48AISI 10450.020.170.180.77AISI 1080

shown in Fig. 1. Eddy Current testing was performed at 27˚C with the fill factor of 0.98. A sinusoidal current was applied to the coil for all samples, primary and secondary voltages (Vx and Vy) and input currents (I) were measured, and the impedance (Z) of the coil was calculated using equation (1) 1).

IVZ /= (1)Calculated impedances of samples were divided by the impedance of the empty coil (Z0) to make a new parameter called normalized impedance (Z/Z0)3, 13).

Fig. 1. General synopsis of the experimental apparatus.

3. Results and DiscussionHughes 14) presents in detail the Eddy Current theory which can be summarized as follows. By passing an alternative current through a coil, fluctuating electromagnetic fields are created. When the sample is introduced into the coil, the electromagnetic fields induce eddy currents, which affect primary and secondary voltages of the coil. These induced variations depend on the Eddy Current magnitude, which in turn, is a function of electrical conductivity and magnetic permeability of the sample as well as test frequency and fill factor (distance between the coils and the sample).The response of eddy current testing is affected by two major parameters. These two parameters are microstructure and residual stress5, 6). The residual stress was kept at a minimum and equal level using the same normalization heat treatment for all samples. Besides, since decarburization depth has an extreme effect on eddy current outputs, surface of all samples were machined to eliminate the decarburized layer. Thus,

the outputs are mainly affected by microstructure. Fig. 2 shows microstructures of four different steels with different pearlite percentage after full annealing treatment. The percentage of pearlite and hardness of the samples are measured by optical observation (using MIP software) and hardness measurement respectively, which are presented in Table 2.

Fig. 2. Metallographic images of AISI a) 1015, b) 1035, c) 1045, d) 1080 steel after full annealing.


30

31


Table 2. The pearlite percentage and carbon content in the steel samples using MIP software and

quantometery, respectively

steel

Percentage of carbonfrom

analysis(%)

Percentage of pearlite

by software(%)

Hardness (RB)

1015 0.13 20.23 651035 0.34 41.35 791045 0.48 64.97 851080 0.77 100 90

By means of regression analysis and by achieving correlation coefficient (R2) for all tested frequencies, the frequency of 650 Hz was chosen as the optimum frequency. The difference in eddy current response of dissimilar microstructures (caused by chemical composition or heat treatment) is as a result of difference in their magnetic properties.In plain carbon steels, as the unequal carbon content is the main cause of difference in pearlite percentage (microstructure), direct relation between eddy current outputs and microstructure will lead to an indirect effect of chemical composition (carbon content) on eddy current outputs. Furthermore, micro-structural changes, or in other words changes in pearlite percentage have a direct effect on hardness of steel samples. As a result, there will be an indirect relation between hardness and eddy current response. Fig. 3 describes these relations.

Fig. 3. Schematic relation between chemical composition, microstructure, hardness and eddy current response 15).

In Fig. 4, relations between eddy current outputs and pearlite percentage of steel are illustrated.As can be seen, increase in pearlite percentage causes eddy current outputs (Vx, Vy, and Z/Z0) to decrease due to difference in their magnetic properties.Regression analysis shows a high accuracy of these relations, particularly for normalized impedance. Thus, for pearlite percentage determination, normalized impedance is the optimum output because of the highest correlation coefficient of 0.99.

Fig 4. The relations between pearlite percentage and eddy current outputs at 650 Hz.

Besides, as is shown in Fig. 5, the same correlation between eddy current outputs and carbon content of the steels can be established. This is due to the direct relation between carbon content and pearlite percentage of steel samples. For these relations again normalized impedance is chosen as the optimum output because of the highest correlation coefficient (R2= 0.98) with respect to the other outputs.

Fig 5. The relation between carbon content of steels and eddy current outputs at 650 Hz.

In the final stage of the current investigation, relations between hardness and the eddy current outputs are studied. These relations are shown in figure 6. As can be seen, the highest correlation coefficient of 0.87 for normalized impedance is a proof of the high ability of this nondestructive method to determine the hardness of steel samples. On the other hand, no suitable relation for primary and secondary voltages can be established. In summary, the results can be used to separate steels with different hardnesses due to different microstructures.


32

Fig 6. The relation between hardness of steels and eddy current outputs at 650 Hz.

Several researches have been performed to investigate the relationship between magnetic hysteresis curve parameters and microstructure of steels 9). The results indicate that an increase in pearlite content of steel causes a coercivity (Hc) increase, and Saturated Magnetic Flux (Bs) decreases. The main effect of increasing pearlite percentage in the microstructure is increasing in magnetic hysteresis loss because of: 1-increasing carbide layers and 2-increasing grain boundaries (due to barrier formation between ferrite and cementite in pearlite lamellar structure). Both of these parameters act as barrier locks and prevent magnetic domains aligning. Therefore, more magnetic field intensity (H) is required to overcome the obstacles against aligning the domains, and therefore more coercivity is needed.Indeed, in all samples, by increasing pearlite percentage and hardness, hysteresis loss will increase and magnetic permeability will decrease. So, considering equation (2), it can be concluded that a decrease in μ results in decrease in self-induction coefficient (L).

lANL /2µ= (2)Where μ is magnetic permeability; N, number of turns round the coil; A, cross section area and l, the coil length.As a result, according to the following equations, by decreasing the magnetic permeability (μ), induction resistance (XL) is decreased. Besides, since in ferromagnetic alloys such as steel, the effect of permeability or reactance is much stronger than that of resistance, impedance (Z) is also decreased.

XL=2πfL (3)22 RXZ L += IV /= (4)

According to equation (4), the impedance decreases

with a increase in the pearlite percentage, hardness and carbon content. The reduction of impedance is a good reason for decreasing the voltage output of eddy current with an increase in the pearlite percentage, hardness and carbon content (Figs. 4, 5 and 6).

4. ConclusionIn the present study, eddy current method was used to determine the pearlite percentage, carbon content and the hardness of steel samples. It was shown that measured (primary and secondary voltages) and calculated (normalized impedance) parameters have good relationship with mentioned micro-structural characteristics. For all samples, the measured and calculated parameters decreased with increasing pearlite percentage, hardness and carbon content. The best relation between the micro-structural characteristics of the samples and eddy current response of the samples can be established using normalized impedance.

References[1] D.C. Jiles: Review of magnetic methods for nondestructive evaluation (Part 2), NDT Int. 23(1990), 83. [2] D.E. Bray and R.K. Stanley: Nondestructive evaluation: a tool design, manufacturing and service, CRC Press, Boca Raton FL, (1997), 415.[3] D.J. Hagemair: Fundamentals of Eddy Current Testing, ASNT, (1990), 270.[4] S. Konoplyuk, T. Abe, T. Uchimoto, T. Takagi, M. Kurosawa: J. NDT&E Int., 38(2005), 623. [5] T. Uchimoto, T. Takagia, S. Konoplyuka, T. Abeb, H. Huanga, M. Kurosawa: J. Magn. Magn. Mater., 259(2003), 493.[6] J. Čech: J. NDT Int., 23(1990), 93.[7] D. Mercier, J. Lesage, X. Decoopman, D. Chicot: NDT & E Int., 39(2006), 652.[8] X.J, Hao, W. Yin, M. Strangwood, A.J. Peyton, P.F. Morris, C.L. Davis: Scripta Mater., 58(2008), 1033.[9] F. Rumiche, J.E. Indacochea, M.L. Wang, J. Mater. Eng. Perform: 17 (2008), 586.[10] J. Degauque, B. Astie, J.L. Porteseil, R. Vergne: J. Magn. Magn. Mater., 26(1982), 261. [11] J. Anglada-Rivera, L.R. Padovese, J. Capo-Sanchez, J. Magn. Magn. Mater: 231(2001), 299.[12] B.K. Tanner, J.A. Szpunar, S.N.M. Willcock, L.L. Morgan, P.A. Mundell, J Mater Sci: 23(1988), 4534.[13] P.J. Shull: Nondestructive evaluation: theory, techniques and applications, CRC Press, New York: Marcel Dekker Inc, (2002), 279.[14] D. E. Hughes, J. Franklin Inst., 116(1883), 128.[15] M. Sheikhamiri, M. Kashefi: NDT & E Int., 42(2009), 618.


Effect of ageing heat treatment on corrosion behaviorof 17-4 PH stainless steel in 3.5% NaCl

M. R. Tavakoli Shoushtari1*

Materials Department, Faculty of Engineering, Shahid Chamran University, Ahvaz 61355, Iran,

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- AbstractThe 17-4PH alloy is a martensitic stainless steel with 3–5 wt% Cu, strengthened by the precipitation hardening. Due to excellent mechanical properties, corrosion resistance and ease of heat treatment, this alloy has unique applications in nuclear power plants and marine constructions. In this paper, the influence of ageing heat treatment, solution annealing followed by ageing at 480, 550 and 620 °C on the corrosion behavior of 17-4PH stainless steel in 3.5% NaCl is reported. Various DC electrochemical measurements and microscopical examination was used. The slow scan rate potentiodynamic polarization revealed that by increasing ageing temperature from 480 to 550 °C, the pitting potential is significantly increased, but further increasing the ageing temperature to 620 °C reduces the pitting potential. Microscopical observation also confirmed the formation of several metastable and stable pits in the sample aged at 620 °C.

Keywords: 17-4PH stainless steel, Corrosion, Ageing heat treatment, Potentiodyanimc polarization.-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

* Corresponding author:Tel: +98 (916) 6022767, Fax:+98 (611) 3336642E-mail: [email protected]: Materials Department, Faculty of Engineering,Shahid Chamran University, Ahvaz 61355, Iran.1. M.Sc., Lecturer

1. IntroductionPrecipitation-hardened stainless steels were first developed during the 1940s, and since then, their application has increased steadily 1,2). The most important of these properties are easy fabrication, high strength, relatively good ductility, and exceptional corrosion resistance 1-5). The 17-4PH (AISI Type 630 or UNS S17400) stainless steel from this family is a martensitic stainless steel containing approximately 3-5 wt.% Cu and is strengthened by the precipitation of nano-dispersed copper enriched zones inside the tempered lath martensitic matrix with low carbon content which are stable at room temperature 3-6). After solution treatment, 17-4PH exhibits a martensitic microstructure but not enough high hardness. Subsequent precipitation ageing treatment in the temperatures between 480 and 620 °C increases hardness and strength 2,4). The application of this alloy has increased in marine constructions, oil and chemical industries and nuclear power plants due to their excellent combination of mechanical property, corrosion and oxidation resistance comparable to type 304 austenitic stainless steel and type 410 martensitic stainless steel, respectively 8,9). But, the alloy exposed to seacoast atmosphere will gradually develop overall light rusting and pitting in all heat-treated conditions. It is almost equal to type 304 and much better than the standard hardenable stainless steels in this

environment 8-10).Researches on corrosion behaviour of 17-4PH stainless steel, particularly in chloride environment, are scarce 11-13). In the present study, the influence of ageing heat treatment on corrosion resistance in 17-4PH stainless steel in 3.5% NaCl solution is investigated by utilizing microscopic studies, determining the pitting parameters by potentiodyanimc polarization measurements, corrosion potential and passivity current.

2. Materials and experimental methods17-4PH stainless steel bar with 93 mm in diameter and 400 mm in length was used. Chemical composition analysis of alloy (in wt%) is 0.01%C, 0.86%Mn, 0.021%P, 0.007%S, 0.8%Si, 15.74%Cr, 3.96%Ni, 0.06%Mo, 2.74%Cu, 0.3%(Nb+V) and Fe balance, which is in agreement with the ASTM A705 (grade 630) standard for precipitation hardening forged stainless steel 14).Three different ageing heat treatments, labeled A, B and C, based on ASTM A 705 14), were performed, as summarized in Table 1, to obtain peak-aged, intermediate and over-aged specimens 15).For corrosion studies, identical samples with cross-section area of 5×5 mm2 were obtained. Before corrosion tests, the samples were degreased in 10% NaOH for 1 minute at 50-60°C, washed by distilled water and dried. In the next step, the samples were ultrasonically cleaned in acetone for 2 minutes at room temperature. The interface between sample and mounting material was covered by lacquer 16), to avoid the crevice corrosion between the specimen and

mounting material during the corrosion tests.

Table 1. Summary of ageing treatment on 17-4PH used in this research.

Label Conditions Heat treatmentABC

H900H1025H1150

Condition A(1h)→480°C×1h→air coolCondition A(1h)→550°C×4h→air coolCondition A(1h)→620°C×4h→air cool

An ACM Potentiostat was employed for the electrochemical tests including working electrode, reference electrode and counter electrode. Saturated calomel reference electrode and platinum wire with a surface of 2 cm2 were chosen as counter electrode. All corrosion tests were performed in a 3.5%wt NaCl solution at ambient temperature. First the rest potential was measured i.e. each sample was placed inside the solution for 3600 seconds and the changes in its corrosion potential were recorded. Each test was repeated 3 times to ensure reproducibility, and average values were reported as rest potential. Potentiodynamic polarization was done with a slow scan rate of 0.05 mV/s in a range of 200 mV cathodic potential, was offset to the rest potential up to the potential value where the sudden current increase was occurred due to extensive pitting corrosion. After anodic polarization close to pitting potential, the samples was ultrasonically cleaned, then they were washed with alcohol and dried. The samples were then etched by Vilella’s reagent and stable and metastable pitting morphologies were observed by SEM. Micrographs were taken by an electron microscope LEO 1455VP (SEM) with a 15 kV power.

3. Results and discussionVariations in OCP of aged specimens of 17-4PH stainless steel immersed in 3.5% NaCl were monitored and the results are presented in Fig. 1. As can be seen, the corrosion potential shows the tendency of passivity characteristics by gradual increasing with time. However, all three samples almost reach the steady state after one hour exposure in 3.5% NaCl. Comparing OCP values reveals that while ageing at 480 °C modifies the OCP to -118 mV; it reduces the OCP to -130 mV by ageing at 620 °C.

Fig. 1. Variation of OCP for aged specimens of 17-4PH stainless steel immersed for 1 hour in 3.5% NaCl solution.

The slow scan rate potentiodynamic polarization curves for aged samples of 17-4PH stainless steel, A, B and C are shown in Fig. 2. The values of OCP (after one hour immersion time), the pitting potential (Epit) and passivity current (ipass) were obtained and the results were shown in Table 2.

Table 2. The values of OCP (after one hour immersion time), the pitting potential (Epit ) and passivity

current (ipass ).Condition OCP(mV) Epit(mV) ipass(µA)

A -118 +173 0.27B -120 +205 0.18C -130 +124 0.1-0.22

Concerning the pitting resistance, it appears that by increasing ageing temperature from 480 to 550 °C, the pitting potential is significantly raised, but further increasing the ageing temperature up to 620 °C reduces the pitting potential down (Table 2). Therefore, ageing at 550 °C, sample B, results the highest pitting resistance. The passivity currents in the range of 0.1 to 0.3 µA/cm2 is obtained for aged samples. By utilizing a very slow scan rate (0.05 mV/s) potentiodynamic polarization, the existence of metastable pits could also be exhibited as the current fluctuations in passivity range (Fig. 2).

Fig. 2. Potentiodynamic polarization curve for 17-4PH stainless steel in labeled A, B and C, immersed in 3.5% NaCl solution. Scan rate was 0.05 mV/s.

In passivity potential domain, the amplitude of fluctuations is bigger in A (aging in 480°C) and B than the one in C, while the fluctuations frequency (the number of current peaks) is greater in C than that in A and B. This means that the pitting sites in C (aging in 620°C) are much more than those in A and B, but they generate less current compared to the A and B. This subject is also confirmed by Fig. 3. Passivity current density curves versus time is extracted from potentiodynamic polarization curves (Fig.2), offset to the anodic potential up to the pitting potential, for aged samples of 17-4PH stainless steel, A, B and C are shown in Fig. 3. The irregular current fluctuations


34


35

observed on curves can be attributed to the formation of metastable pits, which is more enhanced in C. This again confirms the previous potentiodynamic measurements in Fig. 2. The difference in amplitude and frequency of metastable pits can be attributed to the microstructure changes including the volume fraction and the morphology of constituents and precipitates during various ageing heat treatment.

Fig. 3. Passivity current density curves versus time are extracted from potentiodynamic polarization curves, offset to the anodic potential up to the pitting potential, for aged samples of 17-4PH stainless steel, A, B and C.

In summary, potentiodynamic polarization measurements reveal that in B, ageing at 550 °C, leads to the optimum corrosion resistance including higher OCP, pitting potential and passivity potential domain. On the other hand, sample C (ageing at 620 °C) results in the worst corrosion resistance. This is an indication of existence of a marked microstructure difference in C which can be attributed to volume fraction of ferrite, morphology and distribution of copper rich precipitates and the amount of reverted austenite 2). Concerning the influence of reverted austenite, it has been reported that copper precipitates, are expected to be a favorable nucleation site for reversed austenite, since both copper and austenite have the same FCC structure with the similar lattice parameter. During ageing at high temperature above 600 °C, segregation and diffusion of austenite stabilizer’s Cu, Ni and Fe atoms would have a profound influence on the precipitates and the martensite matrix, leading to austenite forming elements enriched areas around the copper precipitates, which could trigger reversed austenite to nucleate. Cr content which is ferrite former conversely decreases the reverted austenite. Therefore, C and N which are also austenite stabilizers may concentrate an the reverted austenite. Moreover, the growth of reversed austenite attracted considerable amounts of Cu and Ni from the martensite matrix, since the solubility of these elements in austenite was

much higher 2,5). The reverted austenite was a stable phase only in the C condition (sample aged at 620 °C) which encompassed the ε-copper rich precipitates. This phenomenon has not been reported in A and B 2). The overall effects of the above parameters are expected to be the reason higher galvanic current was monitored in C compared to A and B. In addition, the presence of white lamellar recrystallized alpha ferrite, which has the highest value in C (both in size and volume fraction) 5), could also be another reason which helps to explain different currents observed in A, B and C. This phase does not exist in A. In A (sample aged at 480 °C), during precipitation of copper, coherent BCC clusters nucleate and by increasing the ageing temperature in B, they grow in the supersaturated BCC matrix, loose slightly their coherency and then by further increasing of ageing temperature in C, they subsequently transform to incoherent FCC ε-copper rich precipitates reaching a certain critical size around 30 nm 2,17). Therefore, coarsening and lowering coherency of precipitates in higher ageing temperature results in the lower corrosion resistance in C. More investigations with accurate techniques such as TEM are needed to determine the microstructure influence and corrosion initiation sites, and to illuminate the exact mechanism.In order to observe morphology of pits, anodic polarization of samples was carried out close to their pitting potential. Formation of several metastable and stable pits in sample aged in A, B and C was observed and the micrographs were shown in Fig. 4.Fig. 4 illustrate stable pits morphology with lacy cover over pits formed at sample aged at 620 °C. Formation of a few metastable pits in other conditions, A and B was also detected.

4. ConclusionThe aim of this work was to study the effect of ageing treatment on the corrosion behavior of 17-4PH stainless steel in 3.5% NaCl. The results can be summarized as follows; The slow scan rate potentiodynamic polarization revealed that by increasing ageing temperature from 480 to 550 °C, the pitting potential is considerably increased, but further rising the ageing temperature up to 620 °C reduces the pitting potential. This is an indication of existence of a marked microstructure difference in sample aged at 620 °C which can be attributed to volume fraction of ferrite, morphology and distribution of copper rich precipitates and the amount of reverted austenite. Microscopical observation after anodic polarization close to pitting potential, also confirmed the formation of several metastable and stable pits in sample aged at 620 °C.


36

Fig. 4. Pits morphology of aged 17-4PH, after anodic polarization close to their pitting potential.

References[1] J.R. Davis, ASM Specialty Handbook Stainless Steel, Volume 34, 1994, ASM International, Materi-als Park.[2] C.N. Hsiao, C.S. Chiou and J.R. Yong: Mater. Chem. Phys., 74 (2002), 132.[3] W. Jun, Z. Hong, L. Cong, Q. Shao-Yu and S. Bao-Luo: Nucl. Eng. Des, 236 (2006), 2531.[4] J.D. Bressan, D.P. Daros, A. Sokolowski, R.A. Mesquita and C.A. Barbosa: J. Mater. Process. Tech-nol., 205, 2008, 353-359.[5] J. Hung Wu, C. Kung Lin: J. Mater. Sci., 38 (2003), 965.[6] M. Murayama, Y. Katayama and K. Hono: Metall. Mater. Trans. A., 30A (1999), 345.[7] W.C. Chiang, C.C. Pu, B.L. Yu and J.K. Wu: Mat-er. Lett., 57 (2003), 2485.[8] AK steel, 17-4PH stainless steel product Data Bul-letin, 2000, AK steel corporation, Middleton.[9] Gui-jiang Li, Jun Wang, Qian Peng, Cong Li, Ying

Wang and Bao-luo Shen: Nucl. Instrum. Meth. B, 266 (2008), 1964.[10] A.J. Sedriks, Corr osion of stainless steel, 2th edn, 1996, John Wiley & Sons, p.124.[11] A. U. Malik, N.A. Siddiqi and I. N. Andijani: De-salination, 97 (1994), 189.[12] J. Nowacki: J. Mater. Process. Technol., 157–158 (2004), 578. [13] M.P. Satish Kumar, P. Bala Srinivasan: Mater. Lett., 62 (2008), 2887.[14] ASTM A 705, Standard specification for age-hardening stainless steel forgings, 1996, Annual Book of ASTM standards.[15] L. W. Tsay, W. C. Lee, R. K. Shiue and J. K. Wu: Corr. Sci., 44 (2002), 2101.[16] 14. C. Garcia, F. Martin, P. de Tiedra, Y. Blanco and M. Lopez: Corr. Sci., 50 (2008), 1184.[17] H. R. Habibi Bajguirani, Mat. Sci. Eng. A, 338 (2002), 142.

37

International Journal of Iron & Steel Society of Iran INSTRUCTIONS FOR AUTHORS

International Journal of Iron & Steel Society of Iran (ISSI) is published semiannually by (ISSI). Original contributions are invited from worldwide ISSI members and non-members.

1.Scope: The scope of the journal extends from the core subject matter of iron and steel to multidisciplinary areas in the science and technology of various materials and processes. The journal provides a medium for the publication of original studies on all aspects of materials and processes including preparation, processing, properties, characterization and application.

2.Category: (1) Regular Article (maximum of ten printed pages): An original article that presents a significant extension of knowledge or understanding and is written in such a way that qualified workers can replicate the key elements on the basis of the information given.

(2) Review: An article of an extensive survey on one particular subject, in which information already published is compiled, analyzed and discussed. Reviews are normally published by invitation. Proposals of suitable subjects by prospective authors are welcome.

(3) Note (maximum of three printed pages): (a) An article on a new finding or interesting aspect of an ongoing study which merits prompt preliminary publication in condensed form, a medium for the presentation of (b) disclosure of new research and techniques, (c) topics, opinions or proposals of interest to the readers and (d) criticisms or additional proofs and interpretations in connection with articles previously published in the society journals.

3.Language: All contributions should be written in English or Persian. The paper should contain an abstract both in English and Persian. However for the authors who are not familiar with Persian, The latter will be prepared by the publisher.

4. Units: The use of SI units is standard. Non SI units approved for use with SI are acceptable.

5. Submission of manuscript: Manuscripts should not be submitted if they have already been published or accepted for publication elsewhere. The original and three copies of a manuscript, both complete with Application Form, synopsis and key words, text, references, list of captions, tables, and figures, should be sent to: The Editorial Board of International Journal of ISSI

The Iron and Steel Society of Iran Science and Technology Sheikh Bahai Park, Isfahan Science and Technology Town,Isfahan University of Technology Boulevard, Isfahan, 84156- 83111, Iran (TeleFax): + 98 311 3932121-24 One set of figures should be of a superior quality for direct reproduction for printing. Papers exceeding the page limits may be returned to the author for condensation prior to reviewing. 6. Reviewing: Every manuscript receives reviewing according to established criteria. 7. Revision of manuscript: In case when the original manuscript is returned to the author for revision, one clear copy of a revised manuscript, together with the original manuscript and a letter explaining the changes made, must be resubmitted within three months. 8. Disk-saved manuscript: To save the printing time and cost, it is desirable for the author to supply the final manuscript of the accepted article in the form of a floppy disk or CD.9. Proofs: The representative author will receive the galley proofs of the paper. No new material may be inserted into the proofs. It is essential that the author returns the proofs before a specified deadline to avoid rescheduling of publication in some later issue. 10. Copyright: The submission of a paper implies that, if accepted for publication, copyright is transferred to the Iron and Steel Society of Iran. The society will not refuse any reasonable request for permission to reproduce a part of the journal. 11. Reprint: No page charge is made. Reprints can be obtained at reasonable prices.

38

GUIDE FOR PREPARATION OF MANUSCRIPT

1. Estimation of length: A journal page consists of approximately 1000 words. Figures are usually reduced to fit into one column of 84 mm width: the largest size of a figure, 110 mm×84 mm, is equivalent to 250 words. 2. Typescript: The typescript must be presented in the order: (1) title page, (2) synopsis and key words (except for Note), (3) text, (4) references, (5) appendices, and (6) list of captions, each of which should start on a new page. The sheet must be numbered consecutively with the title page as page 1. All the sections must be typewritten, double spaced throughout, on one side of A4 paper with ample margins all around.

(1) The title page must contain the title, the full name, affiliation, and mailing address of each author. (2) A synopsis must state briefly and clearly the main object, scope and findings of the work within 250 words. Several key words are required to accompany the synopsis. (3) The text in a regular article must include sufficient details to enable qualified workers to reproduce the results. Extensive literature survey is not necessary. Conclusions are convictions based on the evidence presented. (4) References must be numbered consecutively. Reference numbers in the text should be typed as superscripts with a closing parenthesis, for example, 1),

2,3) and

4-6). List all of the references on a separate

page at the end of the text. Include the names of all the authors with the surnames last. Refer to the following examples for the proper format. 1)Journals Use the standard abbreviations for journal names given in the International Standard ISO 4. Give the volume number, the year of publication and the first page number. [Example] M. Kato, S. Mizoguchi and K. Tsuzaki: ISIJ Int., 40(2000), 543. 2) Conference Proceedings Give the title of the proceedings, the editor’s name if any, the publisher’s name, the place of publication, the year of publication and the page number.[Example] Y. Chino, K. Iwai and S. Asai: Proc. of 3rd Int. Symp. on Electromagnetic Processing of Materials, ISIJ, Tokyo, (2000), 279.3) Books Give the title, the volume number, the editor’s name if any, the publisher’s name, the place of publication, the year of publication and the page number.

[Example] (1) W. C. Leslie: The Physical Metallurgy of Steels, McGraw-Hill, New York, (1981), 621. (2) U. F. Kocks, A. S. Argon and M. F. Ashby: Progress in Materials Science, Vol.19, ed. by B. Chalmers, Pergamon Press, Oxford, (1975), 1. 3. Tables: Tables must not appear in the text but should be prepared on separate sheets. They must have captions and simple column headings. 4. Figures: All graphs, charts, drawings, diagrams, and photographs are to be referred to as Figures and should be numbered consecutively in the order that they are cited in the text. Figures must be photographically reproducible. Each figure must appear on a separate sheet and should be identified by figure number, caption and the representative author’s name. Figure captions must be collected on a separate sheet. Figures are normally reduced in a single column of 84 mm width. All lettering should be legible when reduced to this size. a) Photographs should be supplied as glossy prints and pasted firmly on a hard sheet. When several photographs are to make up one presentation, they should be arranged without leaving margins in between and separately identified as (a), (b), (c)... Magnification must be indicated by means of an inscribed scale. b) Line drawings must be drafted with black ink on white drawing paper. High-quality glossy prints are acceptable. c) Color printing can be arranged, if the reviewers judge it necessary for proper presentation. Authors or their institutions must bear the costs. d) Proper places of insertion should be indicated in the right-hand margin of the text. Classification 1. Ironmaking2. Steelmaking 3. Casting and Solidification 4. Fundamentals of High Temperature Processes 5. Chemical and Physical Analysis 6. Forming Processing and Thermomechanical Treatment 7. Welding and Joining 8. Surface Treatment and Corrosion 9. Transformations and Microstructures 10. Mechanical Properties 11. Physical Properties 12. New Materials and Processes 13. Energy 14. Steel Economics 15. Social and Environmental Engineering 16. Refractories

International Journal of Iron and Steel Society of Iran

Subscription Order Form, 2010

Name: Institution: Mailing Address:

Zip Code: Country: Telephone: Fax: E-mail: Date: Signature:

Please check the appropriate box to receive your annual subscription of International Journal of Iron and Steel Society of Iran

Iran Other Countries □Individual □Rls 50000 □US$ 100 □Institutional □Rls 100000 □US$ 200

Payments should be made via ISSI current account #. 0202831627002 Bank Melli Iran, Isfahan University of Technology Branch, Code: 3187. The receipt should be sent to ISSI Address:

Iron & Steel Society of IranScience and Technology Sheikh Bahai Park, Isfahan Science and Technology Town, Isfahan University of Technology Boulevard, Isfahan, 84156- 83111, IranTeleFax: +98 (311) 3932121-24www.issian.comE-mail: [email protected]

2010 volume 7 number 1 - issiran.com

Documents