The researchers consider emotional divorce as the first stage in the process of divorce, which expresses an enormous marital relationship. The emotional divorce symbolizes the existence of a problem in the healthy and correct relationship between a husband and wife (1). Man is a social being and needs communication with other peers. In this regard, the relationship between couples is more important. Correct communication between couples is one of the key ways of reducing the resolution of differences between spouses. If this does not require proper communication, there is the possibility of providing this important need from a non-family channel (2). One of the means of communication in the world is the Internet and especially social networks. Online social networks have become one of the largest social media technologies that have attracted the audience's potential in social and communication relationships. One cannot ignore the fundamental changes resulting from the advent of these technologies in various personal and social life spheres (3).
Unmanaged use of the Internet and social networks can create problems for families, especially couples, including negative effects on marital relationships and increasing inequalities and rising divorce. It can also increase distrust and lies among family members, reduce verbal and face-to-face communication. One cannot imagine that instruments that affect various social groups do not affect the group in the first meaning of the family (4). In a study in the United Kingdom, it has been shown that about one-third of divorces in this country are related to the Facebook social network (5). The study of the causal and meditational conditions of emotional divorce revealed that the divorce is not a single-agent phenomenon, but among the factors (lack of emotional need, male disregard, communicative problems, uncoordinated thoughts and attitudes of men and women, lack of empathy and companionship of men with women, etc.) in conjunction with them leads to emotional divorce (6). Thus, in this research, the mediating role of attitudes toward betrayal and marital conflicts in the relationship between the use of virtual networks and emotional divorce was investigated.
Individuals always do not follow fixed values and norms in their personal and social lives, but this, like many other things, is subject to change and is the main factor in changing values, attitudes, and communication. The role of communication tools is very effective in changing attitudes and values and injecting norms, and social networks can increase divorce among couples (7). Virtual social networks move and direct a wide range of symbols the norms, values, and ideas of messages take on the thoughts within societies to the point where some scholars believe that virtual social networks transform cultural and social life altogether (2). In the meantime, there is no way that people's attitudes toward betrayal undergoes any changes.
Marital infidelity is, in fact, a neglect of the obligation of the spouse, which creates a degree of emotional and physical intimacy with a non-spouse. Although virtual networks have had many benefits, they have become more and more concern in the area of intercourse, and have trafficked a lot of people. Researchers have divided betrayal into three types of sexual, emotional, and Internet-based content that can both be sexually and emotionally explicit (8).
Conflict in human relationships is common and inevitable. Conflict is a phenomenon that is created in conjunction with love and is inevitable. Conflicts arise when one's actions interfere with the actions of another person, as the two individuals get closer to each other, the conflict of forces increases (9). Norton and Baptist (2014) have shown that the use of social networks is associated with increased emotional distress between husband and wife (10). In addition, numerous studies have shown that the source of many marital conflicts is due to the inappropriateness of the relationship between the couple when the couples rather than the relationship with each other, engage themselves more in social networks and cause inappropriate relationships, causing Robin's wives to communicate (11). The concept of emotional divorce is linked to the issue of disagreement, incompatibility or marital conflicts (12).
Porjebeli and Rezaei, in a study examining the relationship between social network use and family cohesion among Naghadeh families, stated that there is a significant and inverse relationship between the hours of virtual network use and family cohesion (13). In another study on the prediction of narcissism, perception of social interactions and marital conflicts based on the use of social networks the findings showed that there is a significant negative relationship between the use of virtual social networks and perception of social interaction, and there is a significant positive relationship between narcissism and marital conflict (14). Torki and Mohammadi stated that social networks and the virtual world as one of today's agents of socialization have a great effect on divorce and the establishment of inter-marital relationships among couples (15). Several studies have examined the capabilities of the virtual platform for teacher education and professional development. In this paper, suggestions are presented to educate teachers and the use of cyberspace highlighting the importance of education and education institutions and the use of cyberspace by teachers (16). Therefore, the teachers are among the most frequent members of the group who are constantly using virtual networks. Therefore, the married teachers of Mashhad were considered as the statistical community of this research. In the present study, it was tried to investigatethe following hypotheses:
1. Attitudes toward betrayal play a mediating role in the relationship between the use of virtual social networks and emotional divorce in Married Teachers of Mashhad 
2. Marital conflicts play a mediating role in the relationship between the use of virtual social networks and emotional divorce in Married Teachers of Mashhad.
Methods
The method of this research is the structural equation. The statistical population included teachers of the districts of Mashhad in the academic year of 2017-2018. According to statistics published in the statistical-analytical journal this year about 73% of the statistical population is female and 27% are male. Morgan table was used to determine the sample size. According to this table, the sample size was determined by 375 people. A sexual proportional sampling method was used. The questionnaires were distributed randomly among teachers in terms of gender ratio. Data were collected by the researcher in a group. In order to test the data, structural modalities were modeled using AMOS software version 25 and Maximum Likelihood Estimation (ML).
Research tool

1. A questionnaire using online social networks

The questionnaire was developed by Safarpour (17) and has 20 questions. The Likert scale has five degrees and the Cronbach's alpha is 0.03. The total score of the questionnaire is in the range of 20 to 100. Obviously, the higher the score, the greater the person's use of social networks.

2. Emotional divorce questionnaire

The questionnaire was set up by Guttman in 2000 in 24 questions in the form of "yes or no" options. The minimum score obtained by the subject is zero and the maximum score is 24. In this study, the reliability of this scale was obtained using Cronbach's alpha of 0.93. The validity of this questionnaire was confirmed by experts, and factor analysis was used to obtain structural validity. Factor contributions and factor load the questionnaire showed that the load factor of all questions ranged from 0.49 to 0.83, which is a reasonable factor (18).

3. Attitude to betrayal (marital) questionnaire

The Attitudinal Attitude Scale was prepared by Mark Watley in 2006. This scale has 12 terms, each question in the seven range from the very opposite (score one) to very much agree (score 7) (Wattie, 2006). In Iran, this scale has been completed and Cronbach's alpha has been 0.84. The average for this sample is 39/15 (19).

4. Kansas Egman& Associates Marital Conflict Questionnaire (KMCS)

by Kenneth Aegman, Virginia Maxley and Walter R Scham has been comprised of 27 balls used to measure marital conflict in 1985. Scoring the questionnaire as a Likert spectrum is four points. This scale has an excellent inner consistency with an alpha of 0.91 to 0.95 for men and an alpha of 0.88 to 0.95 for women. The scale stability is also very good with 6-month openness correlations for three stages, ranging from 0.64 to 0.96. KMCS scales also have a correlation with a number of other scales, which shows their constructive validity. These scales are FACES-II, empathy scales, respect and coordination, several subscales of marital communication questionnaire, and understanding of the relationship and orientation of the marital goal (20).
Results
The descriptive results of the study showed that of the total 375 cases, 270 (73%) were women and 105 (27%) were male. In terms of duration of the marriage, 131 people (35%) were in the range of 0 -10 years, 165 people (44%) were between 11-21 years and 79 people(21%) were in the range of 22-32 years. The mean and standard deviation of the research’s variables are presented in Table 1.

To investigate the research questions, the preconditions for conducting path analysis including the normal distribution of data, the lack of coherence between variables, the sample size, the distance or relative variables of the research were examined. The skewness and elongation values for all of the research variables are within the range (-2, 2).  Therefore, the assumption that the data are normal is confirmed, and these values are listed in Table 2.

 

The presumption of the existence of a linear relationship between dependent variables and prediction is also based on the correlation table presented is in Table 3.

 

To test the hypothesized model, the variance-covariance matrix was used as input data and, due to the normalization of the data; the maximum likelihood (ML) method was used.
In the following, in Table 4, more precisely, the direct, indirect and total coefficients are presented in two standardized and non-standard levels.
Based on this table, standardized β-standardized coefficients for predicting emotional divorce based on attitude toward betrayal are not significant and small (β = 0.06). This is a detailed statement about the meaningful role of the mediating role of attitudes toward betrayal for predicting emotional divorce based on the use of virtual social networks. While standardized coefficient β for predicting marital conflicts is 0.38 using virtual social networks, the coefficient for predicting emotional divorce based on marital conflicts is 0.54 Accordingly, the role of the mediating meanings of marital conflicts along with the non-perceptible role of attitudes toward betrayal, based on the use of virtual social networks, predicts a total of 20% of the variance of emotional divorce. If the direct role of using virtual social networks in predicting emotional divorce along with the role of intermediate variables is considered, 66% of the variance of emotional divorce is predictable.

 

Table 5 contains the fitness indexes of the hypothesized model. In the appropriate models, the comparative Fit Index (CFI), normed Fit Index (NFI) and the incremental fit index (IFI) are higher than 0.9, the Root Mean Square Error of Approximation (RMSEA) is less than 0.08, the Tucker- Levy (TLI), Relative Fitness Index (RFI), Normal Fitness Index (PNFI), PGFI, and parsimonious goodness of fit index (PRATIO) are also in ratios between 0 and 1. In a model with an appropriate fit, the Chi-2 statistic is not meaningful, or the result of dividing it into degrees of freedom is less than two (Hooman, 2014). Based on this table, the model in most indices has been able to fit the appropriate fit, with the exception of Chi-2 and Root estimation of approximation error variance (RMSEA); therefore, the hypothesized model has a good fit.